Upload INT8-quantized model with calibration

yolo_nas_pose_l_int8.onnx.best.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6be26b88c524b32a7fe99eeaa6166ae80bd428e90d7f45b24317115286e1c80
-size 57269617
+oid sha256:23bee9f331f0da470fb3b6d1e4244127e7648c848e5cf9265fbeea5a15c2204d
+size 57380810

yolo_nas_pose_l_int8.onnx.best.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-19:27:37] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-19:27:37] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-19:27:42] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-20:16:25] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-20:16:25] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-20:16:25] [W] * GPU compute time is unstable, with coefficient of variance = 6.69708%.
-[12/28/2023-20:16:25] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-16:29:12] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-16:29:12] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-16:29:16] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-17:18:11] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-17:18:11] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-17:18:11] [W] * GPU compute time is unstable, with coefficient of variance = 2.14899%.
+[01/04/2024-17:18:11] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_l_int8.onnx.int8.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0437c2287a6331143c0b7181c0ce945e117b29eae0d9755f226919acbe4a8cff
-size 57433967
+oid sha256:f0e0b6cf94d6d65e68c969cd873afeffe71477632d733e3a4f8a9d1d9242fa72
+size 57416931

yolo_nas_pose_l_int8.onnx.int8.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-20:16:35] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-20:16:35] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-20:16:40] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-20:31:04] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-20:31:04] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-20:31:04] [W] * GPU compute time is unstable, with coefficient of variance = 6.47493%.
-[12/28/2023-20:31:04] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-17:18:28] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-17:18:28] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-17:18:32] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-17:33:32] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-17:33:32] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-17:33:32] [W] * GPU compute time is unstable, with coefficient of variance = 9.56431%.
+[01/04/2024-17:33:32] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_m_int8.onnx.best.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:095580091b0635e20541e483c6357dbce9cf2ce19bfab25e1197c375f9ca1807
-size 41724504
+oid sha256:1d812fb42c4b1a75b40065100ad8e4532a011b7d5dd8954655ba54262cceca67
+size 41711541

yolo_nas_pose_m_int8.onnx.best.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-10:38:46] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-10:38:46] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-10:38:50] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-11:17:13] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-11:17:13] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-11:17:13] [W] * GPU compute time is unstable, with coefficient of variance = 6.37849%.
-[12/28/2023-11:17:13] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-15:38:52] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-15:38:52] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-15:38:56] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-16:17:05] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-16:17:05] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-16:17:05] [W] * GPU compute time is unstable, with coefficient of variance = 3.17268%.
+[01/04/2024-16:17:05] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_m_int8.onnx.int8.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:649a3dfa56cfade0026b0d476fab2983d60458937e7431604527c0fefc3c853c
-size 41664431
+oid sha256:bee7381541459b8b03631b7ff27dd33fc54d85b6ef4fdc450591c2062130b3f9
+size 41761906

yolo_nas_pose_m_int8.onnx.int8.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-11:17:20] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-11:17:20] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-11:17:24] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-11:29:15] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-11:29:15] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-11:29:15] [W] * GPU compute time is unstable, with coefficient of variance = 6.95793%.
-[12/28/2023-11:29:15] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-16:17:13] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-16:17:13] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-16:17:16] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-16:29:05] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-16:29:05] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-16:29:05] [W] * GPU compute time is unstable, with coefficient of variance = 3.06633%.
+[01/04/2024-16:29:05] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_n_int8.onnx.best.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:022d94ab6ef2f6c6deba8f1cb3e09149d10b3df3fe17f24559f62f7277c75177
-size 10228093
+oid sha256:034c3a1e3c8f16b6ba305b1b567429a8e42f2ae6c80cb63d074ca8b794844cf3
+size 10229486

yolo_nas_pose_n_int8.onnx.best.engine.err

@@ -1,7 +1,7 @@
-[12/27/2023-20:52:51] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/27/2023-20:52:51] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/27/2023-20:52:54] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/27/2023-21:21:24] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/27/2023-21:21:24] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/27/2023-21:21:24] [W] * GPU compute time is unstable, with coefficient of variance = 3.55467%.
-[12/27/2023-21:21:24] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-14:17:33] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-14:17:33] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-14:17:36] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-14:46:17] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-14:46:17] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-14:46:17] [W] * GPU compute time is unstable, with coefficient of variance = 2.71486%.
+[01/04/2024-14:46:17] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_n_int8.onnx.best.engine.log

@@ -1,323 +1,325 @@
 &&&& RUNNING TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_n_int8.onnx --best --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_n_int8.onnx.best.engine
-[12/27/2023-20:52:47] [I] === Model Options ===
-[12/27/2023-20:52:47] [I] Format: ONNX
-[12/27/2023-20:52:47] [I] Model: yolo_nas_pose_n_int8.onnx
-[12/27/2023-20:52:47] [I] Output:
-[12/27/2023-20:52:47] [I] === Build Options ===
-[12/27/2023-20:52:47] [I] Max batch: explicit batch
-[12/27/2023-20:52:47] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
-[12/27/2023-20:52:47] [I] minTiming: 1
-[12/27/2023-20:52:47] [I] avgTiming: 8
-[12/27/2023-20:52:47] [I] Precision: FP32+FP16+INT8
-[12/27/2023-20:52:47] [I] LayerPrecisions: 
-[12/27/2023-20:52:47] [I] Calibration: Dynamic
-[12/27/2023-20:52:47] [I] Refit: Disabled
-[12/27/2023-20:52:47] [I] Sparsity: Disabled
-[12/27/2023-20:52:47] [I] Safe mode: Disabled
-[12/27/2023-20:52:47] [I] DirectIO mode: Disabled
-[12/27/2023-20:52:47] [I] Restricted mode: Disabled
-[12/27/2023-20:52:47] [I] Build only: Disabled
-[12/27/2023-20:52:47] [I] Save engine: yolo_nas_pose_n_int8.onnx.best.engine
-[12/27/2023-20:52:47] [I] Load engine: 
-[12/27/2023-20:52:47] [I] Profiling verbosity: 0
-[12/27/2023-20:52:47] [I] Tactic sources: Using default tactic sources
-[12/27/2023-20:52:47] [I] timingCacheMode: local
-[12/27/2023-20:52:47] [I] timingCacheFile: 
-[12/27/2023-20:52:47] [I] Heuristic: Disabled
-[12/27/2023-20:52:47] [I] Preview Features: Use default preview flags.
-[12/27/2023-20:52:47] [I] Input(s)s format: fp32:CHW
-[12/27/2023-20:52:47] [I] Output(s)s format: fp32:CHW
-[12/27/2023-20:52:47] [I] Input build shapes: model
-[12/27/2023-20:52:47] [I] Input calibration shapes: model
-[12/27/2023-20:52:47] [I] === System Options ===
-[12/27/2023-20:52:47] [I] Device: 0
-[12/27/2023-20:52:47] [I] DLACore: 
-[12/27/2023-20:52:47] [I] Plugins:
-[12/27/2023-20:52:47] [I] === Inference Options ===
-[12/27/2023-20:52:47] [I] Batch: Explicit
-[12/27/2023-20:52:47] [I] Input inference shapes: model
-[12/27/2023-20:52:47] [I] Iterations: 10
-[12/27/2023-20:52:47] [I] Duration: 15s (+ 200ms warm up)
-[12/27/2023-20:52:47] [I] Sleep time: 0ms
-[12/27/2023-20:52:47] [I] Idle time: 0ms
-[12/27/2023-20:52:47] [I] Streams: 1
-[12/27/2023-20:52:47] [I] ExposeDMA: Disabled
-[12/27/2023-20:52:47] [I] Data transfers: Enabled
-[12/27/2023-20:52:47] [I] Spin-wait: Disabled
-[12/27/2023-20:52:47] [I] Multithreading: Disabled
-[12/27/2023-20:52:47] [I] CUDA Graph: Disabled
-[12/27/2023-20:52:47] [I] Separate profiling: Disabled
-[12/27/2023-20:52:47] [I] Time Deserialize: Disabled
-[12/27/2023-20:52:47] [I] Time Refit: Disabled
-[12/27/2023-20:52:47] [I] NVTX verbosity: 0
-[12/27/2023-20:52:47] [I] Persistent Cache Ratio: 0
-[12/27/2023-20:52:47] [I] Inputs:
-[12/27/2023-20:52:47] [I] === Reporting Options ===
-[12/27/2023-20:52:47] [I] Verbose: Disabled
-[12/27/2023-20:52:47] [I] Averages: 100 inferences
-[12/27/2023-20:52:47] [I] Percentiles: 90,95,99
-[12/27/2023-20:52:47] [I] Dump refittable layers:Disabled
-[12/27/2023-20:52:47] [I] Dump output: Disabled
-[12/27/2023-20:52:47] [I] Profile: Disabled
-[12/27/2023-20:52:47] [I] Export timing to JSON file: 
-[12/27/2023-20:52:47] [I] Export output to JSON file: 
-[12/27/2023-20:52:47] [I] Export profile to JSON file: 
-[12/27/2023-20:52:47] [I] 
-[12/27/2023-20:52:47] [I] === Device Information ===
-[12/27/2023-20:52:47] [I] Selected Device: Orin
-[12/27/2023-20:52:47] [I] Compute Capability: 8.7
-[12/27/2023-20:52:47] [I] SMs: 8
-[12/27/2023-20:52:47] [I] Compute Clock Rate: 0.624 GHz
-[12/27/2023-20:52:47] [I] Device Global Memory: 7471 MiB
-[12/27/2023-20:52:47] [I] Shared Memory per SM: 164 KiB
-[12/27/2023-20:52:47] [I] Memory Bus Width: 128 bits (ECC disabled)
-[12/27/2023-20:52:47] [I] Memory Clock Rate: 0.624 GHz
-[12/27/2023-20:52:47] [I] 
-[12/27/2023-20:52:47] [I] TensorRT version: 8.5.2
-[12/27/2023-20:52:48] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3041 (MiB)
-[12/27/2023-20:52:51] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +283, now: CPU 574, GPU 3346 (MiB)
-[12/27/2023-20:52:51] [I] Start parsing network model
-[12/27/2023-20:52:51] [I] [TRT] ----------------------------------------------------------------
-[12/27/2023-20:52:51] [I] [TRT] Input filename:   yolo_nas_pose_n_int8.onnx
-[12/27/2023-20:52:51] [I] [TRT] ONNX IR version:  0.0.8
-[12/27/2023-20:52:51] [I] [TRT] Opset version:    17
-[12/27/2023-20:52:51] [I] [TRT] Producer name:    pytorch
-[12/27/2023-20:52:51] [I] [TRT] Producer version: 2.1.2
-[12/27/2023-20:52:51] [I] [TRT] Domain:           
-[12/27/2023-20:52:51] [I] [TRT] Model version:    0
-[12/27/2023-20:52:51] [I] [TRT] Doc string:       
-[12/27/2023-20:52:51] [I] [TRT] ----------------------------------------------------------------
-[12/27/2023-20:52:54] [I] Finish parsing network model
-[12/27/2023-20:52:58] [I] [TRT] ---------- Layers Running on DLA ----------
-[12/27/2023-20:52:58] [I] [TRT] ---------- Layers Running on GPU ----------
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1203) [Constant]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1204) [Constant]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1205) [Constant]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 485) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 501) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 548) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 564) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 580) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/Concat_/model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add_output_0_clone_0 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 630) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 646) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 662) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 678) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 719) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 735) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 800) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 816) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/Concat_/model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 865) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 881) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 947) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 988) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/Concat_/model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv || model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1053) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1094) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/Concat_/model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] NMS: batched_nms_238
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1207) [NMS]_1_output[DevicetoShapeHostCopy]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
-[12/27/2023-20:52:58] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
-[12/27/2023-20:53:08] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +373, now: CPU 1168, GPU 3783 (MiB)
-[12/27/2023-20:53:10] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +73, now: CPU 1250, GPU 3856 (MiB)
-[12/27/2023-20:53:10] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
-[12/27/2023-21:20:58] [I] [TRT] Total Activation Memory: 7904134656
-[12/27/2023-21:20:58] [I] [TRT] Detected 1 inputs and 1 output network tensors.
-[12/27/2023-21:21:05] [I] [TRT] Total Host Persistent Memory: 298176
-[12/27/2023-21:21:05] [I] [TRT] Total Device Persistent Memory: 77824
-[12/27/2023-21:21:05] [I] [TRT] Total Scratch Memory: 134217728
-[12/27/2023-21:21:05] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 33 MiB, GPU 154 MiB
-[12/27/2023-21:21:05] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 179 steps to complete.
-[12/27/2023-21:21:05] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 201.747ms to assign 13 blocks to 179 nodes requiring 140788224 bytes.
-[12/27/2023-21:21:05] [I] [TRT] Total Activation Memory: 140788224
-[12/27/2023-21:21:08] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1593, GPU 4908 (MiB)
-[12/27/2023-21:21:08] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +6, GPU +8, now: CPU 6, GPU 8 (MiB)
-[12/27/2023-21:21:08] [I] Engine built in 1700.74 sec.
-[12/27/2023-21:21:09] [I] [TRT] Loaded engine size: 9 MiB
-[12/27/2023-21:21:09] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU -1, now: CPU 1244, GPU 4897 (MiB)
-[12/27/2023-21:21:09] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +7, now: CPU 0, GPU 7 (MiB)
-[12/27/2023-21:21:09] [I] Engine deserialized in 0.189318 sec.
-[12/27/2023-21:21:09] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1245, GPU 4898 (MiB)
-[12/27/2023-21:21:09] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +135, now: CPU 0, GPU 142 (MiB)
-[12/27/2023-21:21:09] [I] Setting persistentCacheLimit to 0 bytes.
-[12/27/2023-21:21:09] [I] Using random values for input onnx::Cast_0
-[12/27/2023-21:21:09] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
-[12/27/2023-21:21:09] [I] Using random values for output graph2_flat_predictions
-[12/27/2023-21:21:09] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
-[12/27/2023-21:21:09] [I] Starting inference
-[12/27/2023-21:21:24] [I] Warmup completed 13 queries over 200 ms
-[12/27/2023-21:21:24] [I] Timing trace has 1215 queries over 15.0321 s
-[12/27/2023-21:21:24] [I] 
-[12/27/2023-21:21:24] [I] === Trace details ===
-[12/27/2023-21:21:24] [I] Trace averages of 100 runs:
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.6305 ms - Host latency: 12.7431 ms (enqueue 12.7033 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.2402 ms - Host latency: 12.3526 ms (enqueue 12.3197 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.256 ms - Host latency: 12.3679 ms (enqueue 12.3361 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1815 ms - Host latency: 12.2935 ms (enqueue 12.2622 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1642 ms - Host latency: 12.276 ms (enqueue 12.2445 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.2964 ms - Host latency: 12.4083 ms (enqueue 12.3727 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1804 ms - Host latency: 12.2922 ms (enqueue 12.2598 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.163 ms - Host latency: 12.2745 ms (enqueue 12.2423 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.173 ms - Host latency: 12.2853 ms (enqueue 12.2536 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1462 ms - Host latency: 12.2579 ms (enqueue 12.2245 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1697 ms - Host latency: 12.281 ms (enqueue 12.2499 ms)
-[12/27/2023-21:21:24] [I] Average on 100 runs - GPU latency: 12.1833 ms - Host latency: 12.2942 ms (enqueue 12.2621 ms)
-[12/27/2023-21:21:24] [I] 
-[12/27/2023-21:21:24] [I] === Performance summary ===
-[12/27/2023-21:21:24] [I] Throughput: 80.8271 qps
-[12/27/2023-21:21:24] [I] Latency: min = 11.5059 ms, max = 19.759 ms, mean = 12.343 ms, median = 12.2217 ms, percentile(90%) = 12.6807 ms, percentile(95%) = 12.8347 ms, percentile(99%) = 14.0815 ms
-[12/27/2023-21:21:24] [I] Enqueue Time: min = 11.4648 ms, max = 21.5256 ms, mean = 12.31 ms, median = 12.1904 ms, percentile(90%) = 12.6436 ms, percentile(95%) = 12.7954 ms, percentile(99%) = 13.7207 ms
-[12/27/2023-21:21:24] [I] H2D Latency: min = 0.0810547 ms, max = 0.117188 ms, mean = 0.0989079 ms, median = 0.0996094 ms, percentile(90%) = 0.100586 ms, percentile(95%) = 0.100586 ms, percentile(99%) = 0.101807 ms
-[12/27/2023-21:21:24] [I] GPU Compute Time: min = 11.4023 ms, max = 19.6559 ms, mean = 12.2311 ms, median = 12.1094 ms, percentile(90%) = 12.564 ms, percentile(95%) = 12.7192 ms, percentile(99%) = 13.9766 ms
-[12/27/2023-21:21:24] [I] D2H Latency: min = 0.00292969 ms, max = 0.0380859 ms, mean = 0.0129526 ms, median = 0.0117188 ms, percentile(90%) = 0.0170898 ms, percentile(95%) = 0.019043 ms, percentile(99%) = 0.027832 ms
-[12/27/2023-21:21:24] [I] Total Host Walltime: 15.0321 s
-[12/27/2023-21:21:24] [I] Total GPU Compute Time: 14.8608 s
-[12/27/2023-21:21:24] [I] Explanations of the performance metrics are printed in the verbose logs.
-[12/27/2023-21:21:24] [I] 
+[01/04/2024-14:17:23] [I] === Model Options ===
+[01/04/2024-14:17:23] [I] Format: ONNX
+[01/04/2024-14:17:23] [I] Model: yolo_nas_pose_n_int8.onnx
+[01/04/2024-14:17:23] [I] Output:
+[01/04/2024-14:17:23] [I] === Build Options ===
+[01/04/2024-14:17:23] [I] Max batch: explicit batch
+[01/04/2024-14:17:23] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
+[01/04/2024-14:17:23] [I] minTiming: 1
+[01/04/2024-14:17:23] [I] avgTiming: 8
+[01/04/2024-14:17:23] [I] Precision: FP32+FP16+INT8
+[01/04/2024-14:17:23] [I] LayerPrecisions: 
+[01/04/2024-14:17:23] [I] Calibration: Dynamic
+[01/04/2024-14:17:23] [I] Refit: Disabled
+[01/04/2024-14:17:23] [I] Sparsity: Disabled
+[01/04/2024-14:17:23] [I] Safe mode: Disabled
+[01/04/2024-14:17:23] [I] DirectIO mode: Disabled
+[01/04/2024-14:17:23] [I] Restricted mode: Disabled
+[01/04/2024-14:17:23] [I] Build only: Disabled
+[01/04/2024-14:17:23] [I] Save engine: yolo_nas_pose_n_int8.onnx.best.engine
+[01/04/2024-14:17:23] [I] Load engine: 
+[01/04/2024-14:17:23] [I] Profiling verbosity: 0
+[01/04/2024-14:17:23] [I] Tactic sources: Using default tactic sources
+[01/04/2024-14:17:23] [I] timingCacheMode: local
+[01/04/2024-14:17:23] [I] timingCacheFile: 
+[01/04/2024-14:17:23] [I] Heuristic: Disabled
+[01/04/2024-14:17:23] [I] Preview Features: Use default preview flags.
+[01/04/2024-14:17:23] [I] Input(s)s format: fp32:CHW
+[01/04/2024-14:17:23] [I] Output(s)s format: fp32:CHW
+[01/04/2024-14:17:23] [I] Input build shapes: model
+[01/04/2024-14:17:23] [I] Input calibration shapes: model
+[01/04/2024-14:17:23] [I] === System Options ===
+[01/04/2024-14:17:23] [I] Device: 0
+[01/04/2024-14:17:23] [I] DLACore: 
+[01/04/2024-14:17:23] [I] Plugins:
+[01/04/2024-14:17:23] [I] === Inference Options ===
+[01/04/2024-14:17:23] [I] Batch: Explicit
+[01/04/2024-14:17:23] [I] Input inference shapes: model
+[01/04/2024-14:17:23] [I] Iterations: 10
+[01/04/2024-14:17:23] [I] Duration: 15s (+ 200ms warm up)
+[01/04/2024-14:17:23] [I] Sleep time: 0ms
+[01/04/2024-14:17:23] [I] Idle time: 0ms
+[01/04/2024-14:17:23] [I] Streams: 1
+[01/04/2024-14:17:23] [I] ExposeDMA: Disabled
+[01/04/2024-14:17:23] [I] Data transfers: Enabled
+[01/04/2024-14:17:23] [I] Spin-wait: Disabled
+[01/04/2024-14:17:23] [I] Multithreading: Disabled
+[01/04/2024-14:17:23] [I] CUDA Graph: Disabled
+[01/04/2024-14:17:23] [I] Separate profiling: Disabled
+[01/04/2024-14:17:23] [I] Time Deserialize: Disabled
+[01/04/2024-14:17:23] [I] Time Refit: Disabled
+[01/04/2024-14:17:23] [I] NVTX verbosity: 0
+[01/04/2024-14:17:23] [I] Persistent Cache Ratio: 0
+[01/04/2024-14:17:23] [I] Inputs:
+[01/04/2024-14:17:23] [I] === Reporting Options ===
+[01/04/2024-14:17:23] [I] Verbose: Disabled
+[01/04/2024-14:17:23] [I] Averages: 100 inferences
+[01/04/2024-14:17:23] [I] Percentiles: 90,95,99
+[01/04/2024-14:17:23] [I] Dump refittable layers:Disabled
+[01/04/2024-14:17:23] [I] Dump output: Disabled
+[01/04/2024-14:17:23] [I] Profile: Disabled
+[01/04/2024-14:17:23] [I] Export timing to JSON file: 
+[01/04/2024-14:17:23] [I] Export output to JSON file: 
+[01/04/2024-14:17:23] [I] Export profile to JSON file: 
+[01/04/2024-14:17:23] [I] 
+[01/04/2024-14:17:24] [I] === Device Information ===
+[01/04/2024-14:17:24] [I] Selected Device: Orin
+[01/04/2024-14:17:24] [I] Compute Capability: 8.7
+[01/04/2024-14:17:24] [I] SMs: 8
+[01/04/2024-14:17:24] [I] Compute Clock Rate: 0.624 GHz
+[01/04/2024-14:17:24] [I] Device Global Memory: 7471 MiB
+[01/04/2024-14:17:24] [I] Shared Memory per SM: 164 KiB
+[01/04/2024-14:17:24] [I] Memory Bus Width: 128 bits (ECC disabled)
+[01/04/2024-14:17:24] [I] Memory Clock Rate: 0.624 GHz
+[01/04/2024-14:17:24] [I] 
+[01/04/2024-14:17:24] [I] TensorRT version: 8.5.2
+[01/04/2024-14:17:29] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 2718 (MiB)
+[01/04/2024-14:17:33] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +435, now: CPU 574, GPU 3215 (MiB)
+[01/04/2024-14:17:33] [I] Start parsing network model
+[01/04/2024-14:17:33] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:17:33] [I] [TRT] Input filename:   yolo_nas_pose_n_int8.onnx
+[01/04/2024-14:17:33] [I] [TRT] ONNX IR version:  0.0.8
+[01/04/2024-14:17:33] [I] [TRT] Opset version:    17
+[01/04/2024-14:17:33] [I] [TRT] Producer name:    pytorch
+[01/04/2024-14:17:33] [I] [TRT] Producer version: 2.1.2
+[01/04/2024-14:17:33] [I] [TRT] Domain:           
+[01/04/2024-14:17:33] [I] [TRT] Model version:    0
+[01/04/2024-14:17:33] [I] [TRT] Doc string:       
+[01/04/2024-14:17:33] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:17:36] [I] Finish parsing network model
+[01/04/2024-14:17:40] [I] [TRT] ---------- Layers Running on DLA ----------
+[01/04/2024-14:17:40] [I] [TRT] ---------- Layers Running on GPU ----------
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1204) [Constant]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1205) [Constant]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1206) [Constant]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 485) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 501) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 548) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 564) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 580) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/Concat_/model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add_output_0_clone_0 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 630) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 646) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 662) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 678) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 719) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 735) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 800) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 816) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/Concat_/model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 865) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 881) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 947) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 988) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/Concat_/model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv || model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1054) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1095) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/Concat_/model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] NMS: batched_nms_238
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1208) [NMS]_1_output[DevicetoShapeHostCopy]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
+[01/04/2024-14:17:40] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
+[01/04/2024-14:17:55] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +815, now: CPU 1168, GPU 4076 (MiB)
+[01/04/2024-14:17:58] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +138, now: CPU 1250, GPU 4214 (MiB)
+[01/04/2024-14:17:58] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
+[01/04/2024-14:45:51] [I] [TRT] Total Activation Memory: 7900558848
+[01/04/2024-14:45:51] [I] [TRT] Detected 1 inputs and 1 output network tensors.
+[01/04/2024-14:45:58] [I] [TRT] Total Host Persistent Memory: 300704
+[01/04/2024-14:45:58] [I] [TRT] Total Device Persistent Memory: 116736
+[01/04/2024-14:45:58] [I] [TRT] Total Scratch Memory: 134217728
+[01/04/2024-14:45:58] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 33 MiB, GPU 154 MiB
+[01/04/2024-14:45:58] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 176 steps to complete.
+[01/04/2024-14:45:58] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 60.5874ms to assign 13 blocks to 176 nodes requiring 140788224 bytes.
+[01/04/2024-14:45:58] [I] [TRT] Total Activation Memory: 140788224
+[01/04/2024-14:46:01] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1593, GPU 5386 (MiB)
+[01/04/2024-14:46:01] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +6, GPU +8, now: CPU 6, GPU 8 (MiB)
+[01/04/2024-14:46:01] [I] Engine built in 1717.31 sec.
+[01/04/2024-14:46:01] [I] [TRT] Loaded engine size: 9 MiB
+[01/04/2024-14:46:02] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1244, GPU 5378 (MiB)
+[01/04/2024-14:46:02] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +7, now: CPU 0, GPU 7 (MiB)
+[01/04/2024-14:46:02] [I] Engine deserialized in 0.18586 sec.
+[01/04/2024-14:46:02] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU -1, now: CPU 1245, GPU 5378 (MiB)
+[01/04/2024-14:46:02] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +135, now: CPU 0, GPU 142 (MiB)
+[01/04/2024-14:46:02] [I] Setting persistentCacheLimit to 0 bytes.
+[01/04/2024-14:46:02] [I] Using random values for input onnx::Cast_0
+[01/04/2024-14:46:02] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
+[01/04/2024-14:46:02] [I] Using random values for output graph2_flat_predictions
+[01/04/2024-14:46:02] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
+[01/04/2024-14:46:02] [I] Starting inference
+[01/04/2024-14:46:17] [I] Warmup completed 3 queries over 200 ms
+[01/04/2024-14:46:17] [I] Timing trace has 1232 queries over 15.0315 s
+[01/04/2024-14:46:17] [I] 
+[01/04/2024-14:46:17] [I] === Trace details ===
+[01/04/2024-14:46:17] [I] Trace averages of 100 runs:
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.29 ms - Host latency: 12.4041 ms (enqueue 12.3732 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0468 ms - Host latency: 12.1594 ms (enqueue 12.1269 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0958 ms - Host latency: 12.2081 ms (enqueue 12.1741 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0328 ms - Host latency: 12.1444 ms (enqueue 12.1128 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0594 ms - Host latency: 12.1718 ms (enqueue 12.1396 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.065 ms - Host latency: 12.1767 ms (enqueue 12.145 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0382 ms - Host latency: 12.1502 ms (enqueue 12.1181 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0485 ms - Host latency: 12.1607 ms (enqueue 12.1285 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0218 ms - Host latency: 12.1333 ms (enqueue 12.1027 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 11.9903 ms - Host latency: 12.1026 ms (enqueue 12.0704 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 11.9893 ms - Host latency: 12.1013 ms (enqueue 12.0701 ms)
+[01/04/2024-14:46:17] [I] Average on 100 runs - GPU latency: 12.0489 ms - Host latency: 12.1609 ms (enqueue 12.1343 ms)
+[01/04/2024-14:46:17] [I] 
+[01/04/2024-14:46:17] [I] === Performance summary ===
+[01/04/2024-14:46:17] [I] Throughput: 81.9611 qps
+[01/04/2024-14:46:17] [I] Latency: min = 11.582 ms, max = 16.703 ms, mean = 12.1721 ms, median = 12.1212 ms, percentile(90%) = 12.4434 ms, percentile(95%) = 12.5928 ms, percentile(99%) = 13.4438 ms
+[01/04/2024-14:46:17] [I] Enqueue Time: min = 11.5547 ms, max = 16.675 ms, mean = 12.1407 ms, median = 12.0908 ms, percentile(90%) = 12.4111 ms, percentile(95%) = 12.5623 ms, percentile(99%) = 13.3818 ms
+[01/04/2024-14:46:17] [I] H2D Latency: min = 0.0830078 ms, max = 0.146088 ms, mean = 0.1 ms, median = 0.0996094 ms, percentile(90%) = 0.10083 ms, percentile(95%) = 0.101562 ms, percentile(99%) = 0.115234 ms
+[01/04/2024-14:46:17] [I] GPU Compute Time: min = 11.4717 ms, max = 16.5476 ms, mean = 12.0599 ms, median = 12.0098 ms, percentile(90%) = 12.332 ms, percentile(95%) = 12.4795 ms, percentile(99%) = 13.3123 ms
+[01/04/2024-14:46:17] [I] D2H Latency: min = 0.00415039 ms, max = 0.0344238 ms, mean = 0.0122015 ms, median = 0.0117188 ms, percentile(90%) = 0.0146484 ms, percentile(95%) = 0.0159912 ms, percentile(99%) = 0.0258789 ms
+[01/04/2024-14:46:17] [I] Total Host Walltime: 15.0315 s
+[01/04/2024-14:46:17] [I] Total GPU Compute Time: 14.8578 s
+[01/04/2024-14:46:17] [I] Explanations of the performance metrics are printed in the verbose logs.
+[01/04/2024-14:46:17] [I] 
 &&&& PASSED TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_n_int8.onnx --best --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_n_int8.onnx.best.engine

yolo_nas_pose_n_int8.onnx.int8.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1235190611d52c7c4bb3e4edb7652062a0e7c08bf3d721d882d72cef0b6eb7cc
-size 10088004
+oid sha256:b19fa98d82cf9039bb036f74f7f3f994127f88c46b7bfab06353cf82e5f8dc09
+size 10115566

yolo_nas_pose_n_int8.onnx.int8.engine.err

@@ -1,7 +1,7 @@
-[12/27/2023-21:21:30] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/27/2023-21:21:30] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/27/2023-21:21:33] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/27/2023-21:30:30] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/27/2023-21:30:30] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/27/2023-21:30:30] [W] * GPU compute time is unstable, with coefficient of variance = 5.79476%.
-[12/27/2023-21:30:30] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-14:46:24] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-14:46:24] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-14:46:27] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-14:55:32] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-14:55:32] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-14:55:32] [W] * GPU compute time is unstable, with coefficient of variance = 4.3012%.
+[01/04/2024-14:55:32] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_n_int8.onnx.int8.engine.log

@@ -1,323 +1,325 @@
 &&&& RUNNING TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_n_int8.onnx --int8 --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_n_int8.onnx.int8.engine
-[12/27/2023-21:21:26] [I] === Model Options ===
-[12/27/2023-21:21:26] [I] Format: ONNX
-[12/27/2023-21:21:26] [I] Model: yolo_nas_pose_n_int8.onnx
-[12/27/2023-21:21:26] [I] Output:
-[12/27/2023-21:21:26] [I] === Build Options ===
-[12/27/2023-21:21:26] [I] Max batch: explicit batch
-[12/27/2023-21:21:26] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
-[12/27/2023-21:21:26] [I] minTiming: 1
-[12/27/2023-21:21:26] [I] avgTiming: 8
-[12/27/2023-21:21:26] [I] Precision: FP32+INT8
-[12/27/2023-21:21:26] [I] LayerPrecisions: 
-[12/27/2023-21:21:26] [I] Calibration: Dynamic
-[12/27/2023-21:21:26] [I] Refit: Disabled
-[12/27/2023-21:21:26] [I] Sparsity: Disabled
-[12/27/2023-21:21:26] [I] Safe mode: Disabled
-[12/27/2023-21:21:26] [I] DirectIO mode: Disabled
-[12/27/2023-21:21:26] [I] Restricted mode: Disabled
-[12/27/2023-21:21:26] [I] Build only: Disabled
-[12/27/2023-21:21:26] [I] Save engine: yolo_nas_pose_n_int8.onnx.int8.engine
-[12/27/2023-21:21:26] [I] Load engine: 
-[12/27/2023-21:21:26] [I] Profiling verbosity: 0
-[12/27/2023-21:21:26] [I] Tactic sources: Using default tactic sources
-[12/27/2023-21:21:26] [I] timingCacheMode: local
-[12/27/2023-21:21:26] [I] timingCacheFile: 
-[12/27/2023-21:21:26] [I] Heuristic: Disabled
-[12/27/2023-21:21:26] [I] Preview Features: Use default preview flags.
-[12/27/2023-21:21:26] [I] Input(s)s format: fp32:CHW
-[12/27/2023-21:21:26] [I] Output(s)s format: fp32:CHW
-[12/27/2023-21:21:26] [I] Input build shapes: model
-[12/27/2023-21:21:26] [I] Input calibration shapes: model
-[12/27/2023-21:21:26] [I] === System Options ===
-[12/27/2023-21:21:26] [I] Device: 0
-[12/27/2023-21:21:26] [I] DLACore: 
-[12/27/2023-21:21:26] [I] Plugins:
-[12/27/2023-21:21:26] [I] === Inference Options ===
-[12/27/2023-21:21:26] [I] Batch: Explicit
-[12/27/2023-21:21:26] [I] Input inference shapes: model
-[12/27/2023-21:21:26] [I] Iterations: 10
-[12/27/2023-21:21:26] [I] Duration: 15s (+ 200ms warm up)
-[12/27/2023-21:21:26] [I] Sleep time: 0ms
-[12/27/2023-21:21:26] [I] Idle time: 0ms
-[12/27/2023-21:21:26] [I] Streams: 1
-[12/27/2023-21:21:26] [I] ExposeDMA: Disabled
-[12/27/2023-21:21:26] [I] Data transfers: Enabled
-[12/27/2023-21:21:26] [I] Spin-wait: Disabled
-[12/27/2023-21:21:26] [I] Multithreading: Disabled
-[12/27/2023-21:21:26] [I] CUDA Graph: Disabled
-[12/27/2023-21:21:26] [I] Separate profiling: Disabled
-[12/27/2023-21:21:26] [I] Time Deserialize: Disabled
-[12/27/2023-21:21:26] [I] Time Refit: Disabled
-[12/27/2023-21:21:26] [I] NVTX verbosity: 0
-[12/27/2023-21:21:26] [I] Persistent Cache Ratio: 0
-[12/27/2023-21:21:26] [I] Inputs:
-[12/27/2023-21:21:26] [I] === Reporting Options ===
-[12/27/2023-21:21:26] [I] Verbose: Disabled
-[12/27/2023-21:21:26] [I] Averages: 100 inferences
-[12/27/2023-21:21:26] [I] Percentiles: 90,95,99
-[12/27/2023-21:21:26] [I] Dump refittable layers:Disabled
-[12/27/2023-21:21:26] [I] Dump output: Disabled
-[12/27/2023-21:21:26] [I] Profile: Disabled
-[12/27/2023-21:21:26] [I] Export timing to JSON file: 
-[12/27/2023-21:21:26] [I] Export output to JSON file: 
-[12/27/2023-21:21:26] [I] Export profile to JSON file: 
-[12/27/2023-21:21:26] [I] 
-[12/27/2023-21:21:26] [I] === Device Information ===
-[12/27/2023-21:21:26] [I] Selected Device: Orin
-[12/27/2023-21:21:26] [I] Compute Capability: 8.7
-[12/27/2023-21:21:26] [I] SMs: 8
-[12/27/2023-21:21:26] [I] Compute Clock Rate: 0.624 GHz
-[12/27/2023-21:21:26] [I] Device Global Memory: 7471 MiB
-[12/27/2023-21:21:26] [I] Shared Memory per SM: 164 KiB
-[12/27/2023-21:21:26] [I] Memory Bus Width: 128 bits (ECC disabled)
-[12/27/2023-21:21:26] [I] Memory Clock Rate: 0.624 GHz
-[12/27/2023-21:21:26] [I] 
-[12/27/2023-21:21:26] [I] TensorRT version: 8.5.2
-[12/27/2023-21:21:27] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3026 (MiB)
-[12/27/2023-21:21:30] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +281, now: CPU 574, GPU 3329 (MiB)
-[12/27/2023-21:21:30] [I] Start parsing network model
-[12/27/2023-21:21:30] [I] [TRT] ----------------------------------------------------------------
-[12/27/2023-21:21:30] [I] [TRT] Input filename:   yolo_nas_pose_n_int8.onnx
-[12/27/2023-21:21:30] [I] [TRT] ONNX IR version:  0.0.8
-[12/27/2023-21:21:30] [I] [TRT] Opset version:    17
-[12/27/2023-21:21:30] [I] [TRT] Producer name:    pytorch
-[12/27/2023-21:21:30] [I] [TRT] Producer version: 2.1.2
-[12/27/2023-21:21:30] [I] [TRT] Domain:           
-[12/27/2023-21:21:30] [I] [TRT] Model version:    0
-[12/27/2023-21:21:30] [I] [TRT] Doc string:       
-[12/27/2023-21:21:30] [I] [TRT] ----------------------------------------------------------------
-[12/27/2023-21:21:32] [I] Finish parsing network model
-[12/27/2023-21:21:32] [I] FP32 and INT8 precisions have been specified - more performance might be enabled by additionally specifying --fp16 or --best
-[12/27/2023-21:21:36] [I] [TRT] ---------- Layers Running on DLA ----------
-[12/27/2023-21:21:36] [I] [TRT] ---------- Layers Running on GPU ----------
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1203) [Constant]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1204) [Constant]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1205) [Constant]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 485) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 501) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 548) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 564) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 580) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/Concat_/model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add_output_0_clone_0 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 630) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 646) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 662) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 678) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 719) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 735) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 800) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 816) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/Concat_/model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 865) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 881) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 947) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 988) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/Concat_/model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv || model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1053) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1094) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/Concat_/model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] NMS: batched_nms_238
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1207) [NMS]_1_output[DevicetoShapeHostCopy]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
-[12/27/2023-21:21:36] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
-[12/27/2023-21:21:37] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +498, now: CPU 1168, GPU 3892 (MiB)
-[12/27/2023-21:21:38] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +61, now: CPU 1250, GPU 3953 (MiB)
-[12/27/2023-21:21:38] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
-[12/27/2023-21:30:13] [I] [TRT] Total Activation Memory: 7918594560
-[12/27/2023-21:30:13] [I] [TRT] Detected 1 inputs and 1 output network tensors.
-[12/27/2023-21:30:14] [I] [TRT] Total Host Persistent Memory: 299584
-[12/27/2023-21:30:14] [I] [TRT] Total Device Persistent Memory: 116736
-[12/27/2023-21:30:14] [I] [TRT] Total Scratch Memory: 134217728
-[12/27/2023-21:30:14] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 33 MiB, GPU 132 MiB
-[12/27/2023-21:30:14] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 180 steps to complete.
-[12/27/2023-21:30:14] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 39.0902ms to assign 13 blocks to 180 nodes requiring 144065024 bytes.
-[12/27/2023-21:30:14] [I] [TRT] Total Activation Memory: 144065024
-[12/27/2023-21:30:14] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1590, GPU 4967 (MiB)
-[12/27/2023-21:30:14] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +6, GPU +8, now: CPU 6, GPU 8 (MiB)
-[12/27/2023-21:30:15] [I] Engine built in 528.25 sec.
-[12/27/2023-21:30:15] [I] [TRT] Loaded engine size: 9 MiB
-[12/27/2023-21:30:15] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1240, GPU 4970 (MiB)
-[12/27/2023-21:30:15] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +7, now: CPU 0, GPU 7 (MiB)
-[12/27/2023-21:30:15] [I] Engine deserialized in 0.110555 sec.
-[12/27/2023-21:30:15] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1241, GPU 4970 (MiB)
-[12/27/2023-21:30:15] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +138, now: CPU 0, GPU 145 (MiB)
-[12/27/2023-21:30:15] [I] Setting persistentCacheLimit to 0 bytes.
-[12/27/2023-21:30:15] [I] Using random values for input onnx::Cast_0
-[12/27/2023-21:30:15] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
-[12/27/2023-21:30:15] [I] Using random values for output graph2_flat_predictions
-[12/27/2023-21:30:15] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
-[12/27/2023-21:30:15] [I] Starting inference
-[12/27/2023-21:30:30] [I] Warmup completed 11 queries over 200 ms
-[12/27/2023-21:30:30] [I] Timing trace has 1116 queries over 15.0271 s
-[12/27/2023-21:30:30] [I] 
-[12/27/2023-21:30:30] [I] === Trace details ===
-[12/27/2023-21:30:30] [I] Trace averages of 100 runs:
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.303 ms - Host latency: 13.4154 ms (enqueue 13.3824 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.1087 ms - Host latency: 13.2205 ms (enqueue 13.189 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.1353 ms - Host latency: 13.2476 ms (enqueue 13.2151 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.1584 ms - Host latency: 13.27 ms (enqueue 13.2379 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.3766 ms - Host latency: 13.4898 ms (enqueue 13.4566 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.1268 ms - Host latency: 13.2384 ms (enqueue 13.2038 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.4482 ms - Host latency: 13.5607 ms (enqueue 13.5189 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.5223 ms - Host latency: 13.637 ms (enqueue 13.5947 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.7048 ms - Host latency: 13.8213 ms (enqueue 13.7771 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.5633 ms - Host latency: 13.6782 ms (enqueue 13.6358 ms)
-[12/27/2023-21:30:30] [I] Average on 100 runs - GPU latency: 13.1192 ms - Host latency: 13.2312 ms (enqueue 13.1992 ms)
-[12/27/2023-21:30:30] [I] 
-[12/27/2023-21:30:30] [I] === Performance summary ===
-[12/27/2023-21:30:30] [I] Throughput: 74.2658 qps
-[12/27/2023-21:30:30] [I] Latency: min = 12.3477 ms, max = 28.8486 ms, mean = 13.4348 ms, median = 13.165 ms, percentile(90%) = 13.8916 ms, percentile(95%) = 13.9805 ms, percentile(99%) = 16.4219 ms
-[12/27/2023-21:30:30] [I] Enqueue Time: min = 12.3198 ms, max = 28.8143 ms, mean = 13.3985 ms, median = 13.1329 ms, percentile(90%) = 13.8477 ms, percentile(95%) = 13.9395 ms, percentile(99%) = 16.3662 ms
-[12/27/2023-21:30:30] [I] H2D Latency: min = 0.0810547 ms, max = 0.185547 ms, mean = 0.0987979 ms, median = 0.0986328 ms, percentile(90%) = 0.100586 ms, percentile(95%) = 0.100586 ms, percentile(99%) = 0.111328 ms
-[12/27/2023-21:30:30] [I] GPU Compute Time: min = 12.2375 ms, max = 28.7328 ms, mean = 13.3217 ms, median = 13.0537 ms, percentile(90%) = 13.7734 ms, percentile(95%) = 13.8643 ms, percentile(99%) = 16.2988 ms
-[12/27/2023-21:30:30] [I] D2H Latency: min = 0.00292969 ms, max = 0.0380859 ms, mean = 0.0142309 ms, median = 0.0126953 ms, percentile(90%) = 0.0195312 ms, percentile(95%) = 0.0214844 ms, percentile(99%) = 0.0302734 ms
-[12/27/2023-21:30:30] [I] Total Host Walltime: 15.0271 s
-[12/27/2023-21:30:30] [I] Total GPU Compute Time: 14.8671 s
-[12/27/2023-21:30:30] [I] Explanations of the performance metrics are printed in the verbose logs.
-[12/27/2023-21:30:30] [I] 
+[01/04/2024-14:46:20] [I] === Model Options ===
+[01/04/2024-14:46:20] [I] Format: ONNX
+[01/04/2024-14:46:20] [I] Model: yolo_nas_pose_n_int8.onnx
+[01/04/2024-14:46:20] [I] Output:
+[01/04/2024-14:46:20] [I] === Build Options ===
+[01/04/2024-14:46:20] [I] Max batch: explicit batch
+[01/04/2024-14:46:20] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
+[01/04/2024-14:46:20] [I] minTiming: 1
+[01/04/2024-14:46:20] [I] avgTiming: 8
+[01/04/2024-14:46:20] [I] Precision: FP32+INT8
+[01/04/2024-14:46:20] [I] LayerPrecisions: 
+[01/04/2024-14:46:20] [I] Calibration: Dynamic
+[01/04/2024-14:46:20] [I] Refit: Disabled
+[01/04/2024-14:46:20] [I] Sparsity: Disabled
+[01/04/2024-14:46:20] [I] Safe mode: Disabled
+[01/04/2024-14:46:20] [I] DirectIO mode: Disabled
+[01/04/2024-14:46:20] [I] Restricted mode: Disabled
+[01/04/2024-14:46:20] [I] Build only: Disabled
+[01/04/2024-14:46:20] [I] Save engine: yolo_nas_pose_n_int8.onnx.int8.engine
+[01/04/2024-14:46:20] [I] Load engine: 
+[01/04/2024-14:46:20] [I] Profiling verbosity: 0
+[01/04/2024-14:46:20] [I] Tactic sources: Using default tactic sources
+[01/04/2024-14:46:20] [I] timingCacheMode: local
+[01/04/2024-14:46:20] [I] timingCacheFile: 
+[01/04/2024-14:46:20] [I] Heuristic: Disabled
+[01/04/2024-14:46:20] [I] Preview Features: Use default preview flags.
+[01/04/2024-14:46:20] [I] Input(s)s format: fp32:CHW
+[01/04/2024-14:46:20] [I] Output(s)s format: fp32:CHW
+[01/04/2024-14:46:20] [I] Input build shapes: model
+[01/04/2024-14:46:20] [I] Input calibration shapes: model
+[01/04/2024-14:46:20] [I] === System Options ===
+[01/04/2024-14:46:20] [I] Device: 0
+[01/04/2024-14:46:20] [I] DLACore: 
+[01/04/2024-14:46:20] [I] Plugins:
+[01/04/2024-14:46:20] [I] === Inference Options ===
+[01/04/2024-14:46:20] [I] Batch: Explicit
+[01/04/2024-14:46:20] [I] Input inference shapes: model
+[01/04/2024-14:46:20] [I] Iterations: 10
+[01/04/2024-14:46:20] [I] Duration: 15s (+ 200ms warm up)
+[01/04/2024-14:46:20] [I] Sleep time: 0ms
+[01/04/2024-14:46:20] [I] Idle time: 0ms
+[01/04/2024-14:46:20] [I] Streams: 1
+[01/04/2024-14:46:20] [I] ExposeDMA: Disabled
+[01/04/2024-14:46:20] [I] Data transfers: Enabled
+[01/04/2024-14:46:20] [I] Spin-wait: Disabled
+[01/04/2024-14:46:20] [I] Multithreading: Disabled
+[01/04/2024-14:46:20] [I] CUDA Graph: Disabled
+[01/04/2024-14:46:20] [I] Separate profiling: Disabled
+[01/04/2024-14:46:20] [I] Time Deserialize: Disabled
+[01/04/2024-14:46:20] [I] Time Refit: Disabled
+[01/04/2024-14:46:20] [I] NVTX verbosity: 0
+[01/04/2024-14:46:20] [I] Persistent Cache Ratio: 0
+[01/04/2024-14:46:20] [I] Inputs:
+[01/04/2024-14:46:20] [I] === Reporting Options ===
+[01/04/2024-14:46:20] [I] Verbose: Disabled
+[01/04/2024-14:46:20] [I] Averages: 100 inferences
+[01/04/2024-14:46:20] [I] Percentiles: 90,95,99
+[01/04/2024-14:46:20] [I] Dump refittable layers:Disabled
+[01/04/2024-14:46:20] [I] Dump output: Disabled
+[01/04/2024-14:46:20] [I] Profile: Disabled
+[01/04/2024-14:46:20] [I] Export timing to JSON file: 
+[01/04/2024-14:46:20] [I] Export output to JSON file: 
+[01/04/2024-14:46:20] [I] Export profile to JSON file: 
+[01/04/2024-14:46:20] [I] 
+[01/04/2024-14:46:20] [I] === Device Information ===
+[01/04/2024-14:46:20] [I] Selected Device: Orin
+[01/04/2024-14:46:20] [I] Compute Capability: 8.7
+[01/04/2024-14:46:20] [I] SMs: 8
+[01/04/2024-14:46:20] [I] Compute Clock Rate: 0.624 GHz
+[01/04/2024-14:46:20] [I] Device Global Memory: 7471 MiB
+[01/04/2024-14:46:20] [I] Shared Memory per SM: 164 KiB
+[01/04/2024-14:46:20] [I] Memory Bus Width: 128 bits (ECC disabled)
+[01/04/2024-14:46:20] [I] Memory Clock Rate: 0.624 GHz
+[01/04/2024-14:46:20] [I] 
+[01/04/2024-14:46:20] [I] TensorRT version: 8.5.2
+[01/04/2024-14:46:20] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3636 (MiB)
+[01/04/2024-14:46:24] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +284, now: CPU 574, GPU 3941 (MiB)
+[01/04/2024-14:46:24] [I] Start parsing network model
+[01/04/2024-14:46:24] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:46:24] [I] [TRT] Input filename:   yolo_nas_pose_n_int8.onnx
+[01/04/2024-14:46:24] [I] [TRT] ONNX IR version:  0.0.8
+[01/04/2024-14:46:24] [I] [TRT] Opset version:    17
+[01/04/2024-14:46:24] [I] [TRT] Producer name:    pytorch
+[01/04/2024-14:46:24] [I] [TRT] Producer version: 2.1.2
+[01/04/2024-14:46:24] [I] [TRT] Domain:           
+[01/04/2024-14:46:24] [I] [TRT] Model version:    0
+[01/04/2024-14:46:24] [I] [TRT] Doc string:       
+[01/04/2024-14:46:24] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:46:27] [I] Finish parsing network model
+[01/04/2024-14:46:27] [I] FP32 and INT8 precisions have been specified - more performance might be enabled by additionally specifying --fp16 or --best
+[01/04/2024-14:46:31] [I] [TRT] ---------- Layers Running on DLA ----------
+[01/04/2024-14:46:31] [I] [TRT] ---------- Layers Running on GPU ----------
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1204) [Constant]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1205) [Constant]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1206) [Constant]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 485) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 501) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 548) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 564) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 580) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/Concat_/model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add_output_0_clone_0 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 630) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 646) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 662) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 678) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 719) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 735) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 800) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 816) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/Concat_/model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 865) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 881) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 947) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 988) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/Concat_/model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv || model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1054) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1095) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/Concat_/model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] NMS: batched_nms_238
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1208) [NMS]_1_output[DevicetoShapeHostCopy]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
+[01/04/2024-14:46:31] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
+[01/04/2024-14:46:36] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +436, now: CPU 1168, GPU 4440 (MiB)
+[01/04/2024-14:46:37] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +43, now: CPU 1250, GPU 4483 (MiB)
+[01/04/2024-14:46:37] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
+[01/04/2024-14:55:15] [I] [TRT] Total Activation Memory: 7920254464
+[01/04/2024-14:55:15] [I] [TRT] Detected 1 inputs and 1 output network tensors.
+[01/04/2024-14:55:15] [I] [TRT] Total Host Persistent Memory: 300608
+[01/04/2024-14:55:15] [I] [TRT] Total Device Persistent Memory: 119296
+[01/04/2024-14:55:15] [I] [TRT] Total Scratch Memory: 134217728
+[01/04/2024-14:55:15] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 33 MiB, GPU 132 MiB
+[01/04/2024-14:55:15] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 185 steps to complete.
+[01/04/2024-14:55:15] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 40.3888ms to assign 13 blocks to 185 nodes requiring 144141824 bytes.
+[01/04/2024-14:55:15] [I] [TRT] Total Activation Memory: 144141824
+[01/04/2024-14:55:16] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1590, GPU 5342 (MiB)
+[01/04/2024-14:55:16] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +6, GPU +8, now: CPU 6, GPU 8 (MiB)
+[01/04/2024-14:55:16] [I] Engine built in 536.559 sec.
+[01/04/2024-14:55:17] [I] [TRT] Loaded engine size: 9 MiB
+[01/04/2024-14:55:17] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1240, GPU 5345 (MiB)
+[01/04/2024-14:55:17] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +7, now: CPU 0, GPU 7 (MiB)
+[01/04/2024-14:55:17] [I] Engine deserialized in 0.120887 sec.
+[01/04/2024-14:55:17] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1241, GPU 5345 (MiB)
+[01/04/2024-14:55:17] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +138, now: CPU 0, GPU 145 (MiB)
+[01/04/2024-14:55:17] [I] Setting persistentCacheLimit to 0 bytes.
+[01/04/2024-14:55:17] [I] Using random values for input onnx::Cast_0
+[01/04/2024-14:55:17] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
+[01/04/2024-14:55:17] [I] Using random values for output graph2_flat_predictions
+[01/04/2024-14:55:17] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
+[01/04/2024-14:55:17] [I] Starting inference
+[01/04/2024-14:55:32] [I] Warmup completed 11 queries over 200 ms
+[01/04/2024-14:55:32] [I] Timing trace has 1108 queries over 15.0289 s
+[01/04/2024-14:55:32] [I] 
+[01/04/2024-14:55:32] [I] === Trace details ===
+[01/04/2024-14:55:32] [I] Trace averages of 100 runs:
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.2196 ms - Host latency: 13.3319 ms (enqueue 13.2998 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.3938 ms - Host latency: 13.5071 ms (enqueue 13.4684 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.2194 ms - Host latency: 13.3318 ms (enqueue 13.2975 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.2789 ms - Host latency: 13.3906 ms (enqueue 13.3589 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.6773 ms - Host latency: 13.7933 ms (enqueue 13.7451 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.7471 ms - Host latency: 13.8643 ms (enqueue 13.8164 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.5782 ms - Host latency: 13.6935 ms (enqueue 13.6539 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.3669 ms - Host latency: 13.4813 ms (enqueue 13.4435 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.2753 ms - Host latency: 13.3888 ms (enqueue 13.3504 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.3138 ms - Host latency: 13.4258 ms (enqueue 13.3842 ms)
+[01/04/2024-14:55:32] [I] Average on 100 runs - GPU latency: 13.511 ms - Host latency: 13.6262 ms (enqueue 13.5945 ms)
+[01/04/2024-14:55:32] [I] 
+[01/04/2024-14:55:32] [I] === Performance summary ===
+[01/04/2024-14:55:32] [I] Throughput: 73.7247 qps
+[01/04/2024-14:55:32] [I] Latency: min = 12.3433 ms, max = 18.3281 ms, mean = 13.5316 ms, median = 13.4639 ms, percentile(90%) = 14.1807 ms, percentile(95%) = 14.3574 ms, percentile(99%) = 15.7993 ms
+[01/04/2024-14:55:32] [I] Enqueue Time: min = 12.313 ms, max = 18.2793 ms, mean = 13.4932 ms, median = 13.4253 ms, percentile(90%) = 14.1348 ms, percentile(95%) = 14.3091 ms, percentile(99%) = 15.7402 ms
+[01/04/2024-14:55:32] [I] H2D Latency: min = 0.0810547 ms, max = 0.114258 ms, mean = 0.0982483 ms, median = 0.0986328 ms, percentile(90%) = 0.0998535 ms, percentile(95%) = 0.100586 ms, percentile(99%) = 0.102539 ms
+[01/04/2024-14:55:32] [I] GPU Compute Time: min = 12.2305 ms, max = 18.21 ms, mean = 13.4177 ms, median = 13.3489 ms, percentile(90%) = 14.0645 ms, percentile(95%) = 14.2349 ms, percentile(99%) = 15.7002 ms
+[01/04/2024-14:55:32] [I] D2H Latency: min = 0.00292969 ms, max = 0.0498047 ms, mean = 0.0156769 ms, median = 0.0146484 ms, percentile(90%) = 0.0205078 ms, percentile(95%) = 0.0230713 ms, percentile(99%) = 0.03125 ms
+[01/04/2024-14:55:32] [I] Total Host Walltime: 15.0289 s
+[01/04/2024-14:55:32] [I] Total GPU Compute Time: 14.8668 s
+[01/04/2024-14:55:32] [I] Explanations of the performance metrics are printed in the verbose logs.
+[01/04/2024-14:55:32] [I] 
 &&&& PASSED TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_n_int8.onnx --int8 --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_n_int8.onnx.int8.engine

yolo_nas_pose_s_int8.onnx.best.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19f0e5f7536fe7f1f802536c4f9de7a56072c2f442043fe76c02813262b89a90
-size 18008336
+oid sha256:24cb45f2e6bbc0f6183888eae7fa29fbe79918082902ae47814c721c3f57bb68
+size 18052506

yolo_nas_pose_s_int8.onnx.best.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-03:03:31] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-03:03:31] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-03:03:34] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-03:36:12] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-03:36:12] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-03:36:12] [W] * GPU compute time is unstable, with coefficient of variance = 3.47567%.
-[12/28/2023-03:36:12] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-14:55:38] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-14:55:38] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-14:55:41] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-15:28:09] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-15:28:10] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-15:28:10] [W] * GPU compute time is unstable, with coefficient of variance = 4.19166%.
+[01/04/2024-15:28:10] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_s_int8.onnx.best.engine.log

@@ -1,323 +1,324 @@
 &&&& RUNNING TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_s_int8.onnx --best --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_s_int8.onnx.best.engine
-[12/28/2023-03:03:27] [I] === Model Options ===
-[12/28/2023-03:03:27] [I] Format: ONNX
-[12/28/2023-03:03:27] [I] Model: yolo_nas_pose_s_int8.onnx
-[12/28/2023-03:03:27] [I] Output:
-[12/28/2023-03:03:27] [I] === Build Options ===
-[12/28/2023-03:03:27] [I] Max batch: explicit batch
-[12/28/2023-03:03:27] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
-[12/28/2023-03:03:27] [I] minTiming: 1
-[12/28/2023-03:03:27] [I] avgTiming: 8
-[12/28/2023-03:03:27] [I] Precision: FP32+FP16+INT8
-[12/28/2023-03:03:27] [I] LayerPrecisions: 
-[12/28/2023-03:03:27] [I] Calibration: Dynamic
-[12/28/2023-03:03:27] [I] Refit: Disabled
-[12/28/2023-03:03:27] [I] Sparsity: Disabled
-[12/28/2023-03:03:27] [I] Safe mode: Disabled
-[12/28/2023-03:03:27] [I] DirectIO mode: Disabled
-[12/28/2023-03:03:27] [I] Restricted mode: Disabled
-[12/28/2023-03:03:27] [I] Build only: Disabled
-[12/28/2023-03:03:27] [I] Save engine: yolo_nas_pose_s_int8.onnx.best.engine
-[12/28/2023-03:03:27] [I] Load engine: 
-[12/28/2023-03:03:27] [I] Profiling verbosity: 0
-[12/28/2023-03:03:27] [I] Tactic sources: Using default tactic sources
-[12/28/2023-03:03:27] [I] timingCacheMode: local
-[12/28/2023-03:03:27] [I] timingCacheFile: 
-[12/28/2023-03:03:27] [I] Heuristic: Disabled
-[12/28/2023-03:03:27] [I] Preview Features: Use default preview flags.
-[12/28/2023-03:03:27] [I] Input(s)s format: fp32:CHW
-[12/28/2023-03:03:27] [I] Output(s)s format: fp32:CHW
-[12/28/2023-03:03:27] [I] Input build shapes: model
-[12/28/2023-03:03:27] [I] Input calibration shapes: model
-[12/28/2023-03:03:27] [I] === System Options ===
-[12/28/2023-03:03:27] [I] Device: 0
-[12/28/2023-03:03:27] [I] DLACore: 
-[12/28/2023-03:03:27] [I] Plugins:
-[12/28/2023-03:03:27] [I] === Inference Options ===
-[12/28/2023-03:03:27] [I] Batch: Explicit
-[12/28/2023-03:03:27] [I] Input inference shapes: model
-[12/28/2023-03:03:27] [I] Iterations: 10
-[12/28/2023-03:03:27] [I] Duration: 15s (+ 200ms warm up)
-[12/28/2023-03:03:27] [I] Sleep time: 0ms
-[12/28/2023-03:03:27] [I] Idle time: 0ms
-[12/28/2023-03:03:27] [I] Streams: 1
-[12/28/2023-03:03:27] [I] ExposeDMA: Disabled
-[12/28/2023-03:03:27] [I] Data transfers: Enabled
-[12/28/2023-03:03:27] [I] Spin-wait: Disabled
-[12/28/2023-03:03:27] [I] Multithreading: Disabled
-[12/28/2023-03:03:27] [I] CUDA Graph: Disabled
-[12/28/2023-03:03:27] [I] Separate profiling: Disabled
-[12/28/2023-03:03:27] [I] Time Deserialize: Disabled
-[12/28/2023-03:03:27] [I] Time Refit: Disabled
-[12/28/2023-03:03:27] [I] NVTX verbosity: 0
-[12/28/2023-03:03:27] [I] Persistent Cache Ratio: 0
-[12/28/2023-03:03:27] [I] Inputs:
-[12/28/2023-03:03:27] [I] === Reporting Options ===
-[12/28/2023-03:03:27] [I] Verbose: Disabled
-[12/28/2023-03:03:27] [I] Averages: 100 inferences
-[12/28/2023-03:03:27] [I] Percentiles: 90,95,99
-[12/28/2023-03:03:27] [I] Dump refittable layers:Disabled
-[12/28/2023-03:03:27] [I] Dump output: Disabled
-[12/28/2023-03:03:27] [I] Profile: Disabled
-[12/28/2023-03:03:27] [I] Export timing to JSON file: 
-[12/28/2023-03:03:27] [I] Export output to JSON file: 
-[12/28/2023-03:03:27] [I] Export profile to JSON file: 
-[12/28/2023-03:03:27] [I] 
-[12/28/2023-03:03:27] [I] === Device Information ===
-[12/28/2023-03:03:27] [I] Selected Device: Orin
-[12/28/2023-03:03:27] [I] Compute Capability: 8.7
-[12/28/2023-03:03:27] [I] SMs: 8
-[12/28/2023-03:03:27] [I] Compute Clock Rate: 0.624 GHz
-[12/28/2023-03:03:27] [I] Device Global Memory: 7471 MiB
-[12/28/2023-03:03:27] [I] Shared Memory per SM: 164 KiB
-[12/28/2023-03:03:27] [I] Memory Bus Width: 128 bits (ECC disabled)
-[12/28/2023-03:03:27] [I] Memory Clock Rate: 0.624 GHz
-[12/28/2023-03:03:27] [I] 
-[12/28/2023-03:03:27] [I] TensorRT version: 8.5.2
-[12/28/2023-03:03:28] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3018 (MiB)
-[12/28/2023-03:03:31] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +284, now: CPU 574, GPU 3324 (MiB)
-[12/28/2023-03:03:31] [I] Start parsing network model
-[12/28/2023-03:03:31] [I] [TRT] ----------------------------------------------------------------
-[12/28/2023-03:03:31] [I] [TRT] Input filename:   yolo_nas_pose_s_int8.onnx
-[12/28/2023-03:03:31] [I] [TRT] ONNX IR version:  0.0.8
-[12/28/2023-03:03:31] [I] [TRT] Opset version:    17
-[12/28/2023-03:03:31] [I] [TRT] Producer name:    pytorch
-[12/28/2023-03:03:31] [I] [TRT] Producer version: 2.1.2
-[12/28/2023-03:03:31] [I] [TRT] Domain:           
-[12/28/2023-03:03:31] [I] [TRT] Model version:    0
-[12/28/2023-03:03:31] [I] [TRT] Doc string:       
-[12/28/2023-03:03:31] [I] [TRT] ----------------------------------------------------------------
-[12/28/2023-03:03:34] [I] Finish parsing network model
-[12/28/2023-03:03:38] [I] [TRT] ---------- Layers Running on DLA ----------
-[12/28/2023-03:03:38] [I] [TRT] ---------- Layers Running on GPU ----------
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1228) [Constant]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1229) [Constant]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1230) [Constant]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 494) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 510) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 557) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 573) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 589) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 639) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 655) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 671) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 687) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.4.alpha + (Unnamed Layer* 703) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 744) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 760) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 825) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 841) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 890) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 906) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/Concat_/model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 972) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1013) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv || model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1078) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1119) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] NMS: batched_nms_243
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1232) [NMS]_1_output[DevicetoShapeHostCopy]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
-[12/28/2023-03:03:38] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
-[12/28/2023-03:03:51] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +341, now: CPU 1233, GPU 3794 (MiB)
-[12/28/2023-03:03:53] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +76, now: CPU 1315, GPU 3870 (MiB)
-[12/28/2023-03:03:53] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
-[12/28/2023-03:35:46] [I] [TRT] Total Activation Memory: 7917157888
-[12/28/2023-03:35:46] [I] [TRT] Detected 1 inputs and 1 output network tensors.
-[12/28/2023-03:35:53] [I] [TRT] Total Host Persistent Memory: 308256
-[12/28/2023-03:35:53] [I] [TRT] Total Device Persistent Memory: 38912
-[12/28/2023-03:35:53] [I] [TRT] Total Scratch Memory: 134217728
-[12/28/2023-03:35:53] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 74 MiB, GPU 154 MiB
-[12/28/2023-03:35:53] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 164 steps to complete.
-[12/28/2023-03:35:53] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 59.5318ms to assign 13 blocks to 164 nodes requiring 142029824 bytes.
-[12/28/2023-03:35:53] [I] [TRT] Total Activation Memory: 142029824
-[12/28/2023-03:35:56] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1667, GPU 5515 (MiB)
-[12/28/2023-03:35:56] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +15, GPU +16, now: CPU 15, GPU 16 (MiB)
-[12/28/2023-03:35:56] [I] Engine built in 1949.31 sec.
-[12/28/2023-03:35:57] [I] [TRT] Loaded engine size: 17 MiB
-[12/28/2023-03:35:57] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1252, GPU 5486 (MiB)
-[12/28/2023-03:35:57] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +15, now: CPU 0, GPU 15 (MiB)
-[12/28/2023-03:35:57] [I] Engine deserialized in 0.208393 sec.
-[12/28/2023-03:35:57] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1252, GPU 5486 (MiB)
-[12/28/2023-03:35:57] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +135, now: CPU 0, GPU 150 (MiB)
-[12/28/2023-03:35:57] [I] Setting persistentCacheLimit to 0 bytes.
-[12/28/2023-03:35:57] [I] Using random values for input onnx::Cast_0
-[12/28/2023-03:35:57] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
-[12/28/2023-03:35:57] [I] Using random values for output graph2_flat_predictions
-[12/28/2023-03:35:57] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
-[12/28/2023-03:35:57] [I] Starting inference
-[12/28/2023-03:36:12] [I] Warmup completed 11 queries over 200 ms
-[12/28/2023-03:36:12] [I] Timing trace has 1116 queries over 15.0304 s
-[12/28/2023-03:36:12] [I] 
-[12/28/2023-03:36:12] [I] === Trace details ===
-[12/28/2023-03:36:12] [I] Trace averages of 100 runs:
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.3509 ms - Host latency: 13.4657 ms (enqueue 13.4291 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.6114 ms - Host latency: 13.7267 ms (enqueue 13.6814 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2966 ms - Host latency: 13.4103 ms (enqueue 13.3777 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.3229 ms - Host latency: 13.4369 ms (enqueue 13.4032 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.3265 ms - Host latency: 13.4397 ms (enqueue 13.4066 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.3048 ms - Host latency: 13.4183 ms (enqueue 13.3859 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2897 ms - Host latency: 13.4033 ms (enqueue 13.375 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2914 ms - Host latency: 13.4051 ms (enqueue 13.3713 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2786 ms - Host latency: 13.3924 ms (enqueue 13.3595 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2361 ms - Host latency: 13.3498 ms (enqueue 13.3173 ms)
-[12/28/2023-03:36:12] [I] Average on 100 runs - GPU latency: 13.2792 ms - Host latency: 13.3929 ms (enqueue 13.3575 ms)
-[12/28/2023-03:36:12] [I] 
-[12/28/2023-03:36:12] [I] === Performance summary ===
-[12/28/2023-03:36:12] [I] Throughput: 74.2494 qps
-[12/28/2023-03:36:12] [I] Latency: min = 12.5967 ms, max = 18.459 ms, mean = 13.4387 ms, median = 13.2627 ms, percentile(90%) = 13.9451 ms, percentile(95%) = 14.0898 ms, percentile(99%) = 15.3137 ms
-[12/28/2023-03:36:12] [I] Enqueue Time: min = 12.5688 ms, max = 18.406 ms, mean = 13.4045 ms, median = 13.2305 ms, percentile(90%) = 13.9082 ms, percentile(95%) = 14.0646 ms, percentile(99%) = 14.8955 ms
-[12/28/2023-03:36:12] [I] H2D Latency: min = 0.0820312 ms, max = 0.120605 ms, mean = 0.100692 ms, median = 0.101074 ms, percentile(90%) = 0.102295 ms, percentile(95%) = 0.102539 ms, percentile(99%) = 0.113281 ms
-[12/28/2023-03:36:12] [I] GPU Compute Time: min = 12.4849 ms, max = 18.3318 ms, mean = 13.3247 ms, median = 13.1485 ms, percentile(90%) = 13.8306 ms, percentile(95%) = 13.973 ms, percentile(99%) = 15.1929 ms
-[12/28/2023-03:36:12] [I] D2H Latency: min = 0.00292969 ms, max = 0.0505371 ms, mean = 0.0132362 ms, median = 0.0117188 ms, percentile(90%) = 0.0186768 ms, percentile(95%) = 0.0222168 ms, percentile(99%) = 0.0280762 ms
-[12/28/2023-03:36:12] [I] Total Host Walltime: 15.0304 s
-[12/28/2023-03:36:12] [I] Total GPU Compute Time: 14.8704 s
-[12/28/2023-03:36:12] [I] Explanations of the performance metrics are printed in the verbose logs.
-[12/28/2023-03:36:12] [I] 
+[01/04/2024-14:55:33] [I] === Model Options ===
+[01/04/2024-14:55:33] [I] Format: ONNX
+[01/04/2024-14:55:33] [I] Model: yolo_nas_pose_s_int8.onnx
+[01/04/2024-14:55:33] [I] Output:
+[01/04/2024-14:55:33] [I] === Build Options ===
+[01/04/2024-14:55:33] [I] Max batch: explicit batch
+[01/04/2024-14:55:33] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
+[01/04/2024-14:55:33] [I] minTiming: 1
+[01/04/2024-14:55:33] [I] avgTiming: 8
+[01/04/2024-14:55:33] [I] Precision: FP32+FP16+INT8
+[01/04/2024-14:55:33] [I] LayerPrecisions: 
+[01/04/2024-14:55:33] [I] Calibration: Dynamic
+[01/04/2024-14:55:33] [I] Refit: Disabled
+[01/04/2024-14:55:33] [I] Sparsity: Disabled
+[01/04/2024-14:55:33] [I] Safe mode: Disabled
+[01/04/2024-14:55:33] [I] DirectIO mode: Disabled
+[01/04/2024-14:55:33] [I] Restricted mode: Disabled
+[01/04/2024-14:55:33] [I] Build only: Disabled
+[01/04/2024-14:55:33] [I] Save engine: yolo_nas_pose_s_int8.onnx.best.engine
+[01/04/2024-14:55:33] [I] Load engine: 
+[01/04/2024-14:55:33] [I] Profiling verbosity: 0
+[01/04/2024-14:55:33] [I] Tactic sources: Using default tactic sources
+[01/04/2024-14:55:33] [I] timingCacheMode: local
+[01/04/2024-14:55:33] [I] timingCacheFile: 
+[01/04/2024-14:55:33] [I] Heuristic: Disabled
+[01/04/2024-14:55:33] [I] Preview Features: Use default preview flags.
+[01/04/2024-14:55:33] [I] Input(s)s format: fp32:CHW
+[01/04/2024-14:55:33] [I] Output(s)s format: fp32:CHW
+[01/04/2024-14:55:33] [I] Input build shapes: model
+[01/04/2024-14:55:33] [I] Input calibration shapes: model
+[01/04/2024-14:55:33] [I] === System Options ===
+[01/04/2024-14:55:33] [I] Device: 0
+[01/04/2024-14:55:33] [I] DLACore: 
+[01/04/2024-14:55:33] [I] Plugins:
+[01/04/2024-14:55:33] [I] === Inference Options ===
+[01/04/2024-14:55:33] [I] Batch: Explicit
+[01/04/2024-14:55:33] [I] Input inference shapes: model
+[01/04/2024-14:55:33] [I] Iterations: 10
+[01/04/2024-14:55:33] [I] Duration: 15s (+ 200ms warm up)
+[01/04/2024-14:55:33] [I] Sleep time: 0ms
+[01/04/2024-14:55:33] [I] Idle time: 0ms
+[01/04/2024-14:55:33] [I] Streams: 1
+[01/04/2024-14:55:33] [I] ExposeDMA: Disabled
+[01/04/2024-14:55:33] [I] Data transfers: Enabled
+[01/04/2024-14:55:33] [I] Spin-wait: Disabled
+[01/04/2024-14:55:33] [I] Multithreading: Disabled
+[01/04/2024-14:55:33] [I] CUDA Graph: Disabled
+[01/04/2024-14:55:33] [I] Separate profiling: Disabled
+[01/04/2024-14:55:33] [I] Time Deserialize: Disabled
+[01/04/2024-14:55:33] [I] Time Refit: Disabled
+[01/04/2024-14:55:33] [I] NVTX verbosity: 0
+[01/04/2024-14:55:33] [I] Persistent Cache Ratio: 0
+[01/04/2024-14:55:33] [I] Inputs:
+[01/04/2024-14:55:33] [I] === Reporting Options ===
+[01/04/2024-14:55:33] [I] Verbose: Disabled
+[01/04/2024-14:55:33] [I] Averages: 100 inferences
+[01/04/2024-14:55:33] [I] Percentiles: 90,95,99
+[01/04/2024-14:55:33] [I] Dump refittable layers:Disabled
+[01/04/2024-14:55:33] [I] Dump output: Disabled
+[01/04/2024-14:55:33] [I] Profile: Disabled
+[01/04/2024-14:55:33] [I] Export timing to JSON file: 
+[01/04/2024-14:55:33] [I] Export output to JSON file: 
+[01/04/2024-14:55:33] [I] Export profile to JSON file: 
+[01/04/2024-14:55:33] [I] 
+[01/04/2024-14:55:33] [I] === Device Information ===
+[01/04/2024-14:55:33] [I] Selected Device: Orin
+[01/04/2024-14:55:33] [I] Compute Capability: 8.7
+[01/04/2024-14:55:33] [I] SMs: 8
+[01/04/2024-14:55:33] [I] Compute Clock Rate: 0.624 GHz
+[01/04/2024-14:55:33] [I] Device Global Memory: 7471 MiB
+[01/04/2024-14:55:33] [I] Shared Memory per SM: 164 KiB
+[01/04/2024-14:55:33] [I] Memory Bus Width: 128 bits (ECC disabled)
+[01/04/2024-14:55:33] [I] Memory Clock Rate: 0.624 GHz
+[01/04/2024-14:55:33] [I] 
+[01/04/2024-14:55:33] [I] TensorRT version: 8.5.2
+[01/04/2024-14:55:34] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3779 (MiB)
+[01/04/2024-14:55:37] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +284, now: CPU 574, GPU 4083 (MiB)
+[01/04/2024-14:55:37] [I] Start parsing network model
+[01/04/2024-14:55:38] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:55:38] [I] [TRT] Input filename:   yolo_nas_pose_s_int8.onnx
+[01/04/2024-14:55:38] [I] [TRT] ONNX IR version:  0.0.8
+[01/04/2024-14:55:38] [I] [TRT] Opset version:    17
+[01/04/2024-14:55:38] [I] [TRT] Producer name:    pytorch
+[01/04/2024-14:55:38] [I] [TRT] Producer version: 2.1.2
+[01/04/2024-14:55:38] [I] [TRT] Domain:           
+[01/04/2024-14:55:38] [I] [TRT] Model version:    0
+[01/04/2024-14:55:38] [I] [TRT] Doc string:       
+[01/04/2024-14:55:38] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-14:55:41] [I] Finish parsing network model
+[01/04/2024-14:55:45] [I] [TRT] ---------- Layers Running on DLA ----------
+[01/04/2024-14:55:45] [I] [TRT] ---------- Layers Running on GPU ----------
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1229) [Constant]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1230) [Constant]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1231) [Constant]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 494) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 510) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 557) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 573) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 589) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 639) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 655) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 671) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 687) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.4.alpha + (Unnamed Layer* 703) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 744) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 760) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 825) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 841) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 890) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 906) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/Concat_/model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 972) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1013) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv || model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1079) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1120) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] NMS: batched_nms_243
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1233) [NMS]_1_output[DevicetoShapeHostCopy]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
+[01/04/2024-14:55:45] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
+[01/04/2024-14:55:46] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +459, now: CPU 1233, GPU 4661 (MiB)
+[01/04/2024-14:55:46] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +38, now: CPU 1315, GPU 4699 (MiB)
+[01/04/2024-14:55:46] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
+[01/04/2024-15:27:42] [I] [TRT] Total Activation Memory: 7917384192
+[01/04/2024-15:27:42] [I] [TRT] Detected 1 inputs and 1 output network tensors.
+[01/04/2024-15:27:50] [I] [TRT] Total Host Persistent Memory: 308448
+[01/04/2024-15:27:50] [I] [TRT] Total Device Persistent Memory: 653824
+[01/04/2024-15:27:50] [I] [TRT] Total Scratch Memory: 134217728
+[01/04/2024-15:27:50] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 74 MiB, GPU 154 MiB
+[01/04/2024-15:27:50] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 166 steps to complete.
+[01/04/2024-15:27:50] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 59.522ms to assign 13 blocks to 166 nodes requiring 141982720 bytes.
+[01/04/2024-15:27:50] [I] [TRT] Total Activation Memory: 141982720
+[01/04/2024-15:27:53] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1667, GPU 5744 (MiB)
+[01/04/2024-15:27:53] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +15, GPU +16, now: CPU 15, GPU 16 (MiB)
+[01/04/2024-15:27:53] [I] Engine built in 1940.22 sec.
+[01/04/2024-15:27:54] [I] [TRT] Loaded engine size: 17 MiB
+[01/04/2024-15:27:54] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1252, GPU 5595 (MiB)
+[01/04/2024-15:27:54] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +15, now: CPU 0, GPU 15 (MiB)
+[01/04/2024-15:27:54] [I] Engine deserialized in 0.210553 sec.
+[01/04/2024-15:27:54] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU -1, now: CPU 1252, GPU 5594 (MiB)
+[01/04/2024-15:27:54] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +136, now: CPU 0, GPU 151 (MiB)
+[01/04/2024-15:27:54] [I] Setting persistentCacheLimit to 0 bytes.
+[01/04/2024-15:27:54] [I] Using random values for input onnx::Cast_0
+[01/04/2024-15:27:54] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
+[01/04/2024-15:27:54] [I] Using random values for output graph2_flat_predictions
+[01/04/2024-15:27:54] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
+[01/04/2024-15:27:54] [I] Starting inference
+[01/04/2024-15:28:09] [I] Warmup completed 12 queries over 200 ms
+[01/04/2024-15:28:09] [I] Timing trace has 1074 queries over 15.0266 s
+[01/04/2024-15:28:09] [I] 
+[01/04/2024-15:28:09] [I] === Trace details ===
+[01/04/2024-15:28:09] [I] Trace averages of 100 runs:
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.6253 ms - Host latency: 13.7361 ms (enqueue 13.703 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.9431 ms - Host latency: 14.0566 ms (enqueue 14.0098 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.8369 ms - Host latency: 13.9494 ms (enqueue 13.9083 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.8257 ms - Host latency: 13.9381 ms (enqueue 13.8989 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.6064 ms - Host latency: 13.7172 ms (enqueue 13.6832 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 14.264 ms - Host latency: 14.3781 ms (enqueue 14.3258 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.6034 ms - Host latency: 13.7146 ms (enqueue 13.682 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 14.1877 ms - Host latency: 14.3027 ms (enqueue 14.2525 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.7484 ms - Host latency: 13.8601 ms (enqueue 13.8257 ms)
+[01/04/2024-15:28:09] [I] Average on 100 runs - GPU latency: 13.7575 ms - Host latency: 13.8697 ms (enqueue 13.8349 ms)
+[01/04/2024-15:28:09] [I] 
+[01/04/2024-15:28:09] [I] === Performance summary ===
+[01/04/2024-15:28:09] [I] Throughput: 71.4732 qps
+[01/04/2024-15:28:09] [I] Latency: min = 13.0068 ms, max = 17.7432 ms, mean = 13.9607 ms, median = 13.9542 ms, percentile(90%) = 14.7441 ms, percentile(95%) = 14.9595 ms, percentile(99%) = 15.5879 ms
+[01/04/2024-15:28:09] [I] Enqueue Time: min = 12.9634 ms, max = 18.0693 ms, mean = 13.9208 ms, median = 13.9097 ms, percentile(90%) = 14.6982 ms, percentile(95%) = 14.8809 ms, percentile(99%) = 15.5361 ms
+[01/04/2024-15:28:09] [I] H2D Latency: min = 0.0809937 ms, max = 0.114258 ms, mean = 0.0973303 ms, median = 0.0976562 ms, percentile(90%) = 0.0991211 ms, percentile(95%) = 0.0996094 ms, percentile(99%) = 0.101562 ms
+[01/04/2024-15:28:09] [I] GPU Compute Time: min = 12.8984 ms, max = 17.6377 ms, mean = 13.8482 ms, median = 13.8396 ms, percentile(90%) = 14.6279 ms, percentile(95%) = 14.8496 ms, percentile(99%) = 15.4727 ms
+[01/04/2024-15:28:09] [I] D2H Latency: min = 0.00390625 ms, max = 0.0466309 ms, mean = 0.0151338 ms, median = 0.0130615 ms, percentile(90%) = 0.0224609 ms, percentile(95%) = 0.0244141 ms, percentile(99%) = 0.03125 ms
+[01/04/2024-15:28:09] [I] Total Host Walltime: 15.0266 s
+[01/04/2024-15:28:09] [I] Total GPU Compute Time: 14.873 s
+[01/04/2024-15:28:10] [I] Explanations of the performance metrics are printed in the verbose logs.
+[01/04/2024-15:28:10] [I] 
 &&&& PASSED TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_s_int8.onnx --best --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_s_int8.onnx.best.engine

yolo_nas_pose_s_int8.onnx.int8.engine

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b03544a46558d98432ec484cef728f0be750bcbb79762856b78125143998b2d
-size 17958627
+oid sha256:8795eb441b2790005ae4c651d93cda57424a970bde3e5a5bceff132b34cf7c78
+size 17990796

yolo_nas_pose_s_int8.onnx.int8.engine.err

@@ -1,7 +1,7 @@
-[12/28/2023-03:36:18] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
-[12/28/2023-03:36:18] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
-[12/28/2023-03:36:21] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
-[12/28/2023-03:46:46] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
-[12/28/2023-03:46:46] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
-[12/28/2023-03:46:46] [W] * GPU compute time is unstable, with coefficient of variance = 3.57305%.
-[12/28/2023-03:46:46] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.
+[01/04/2024-15:28:15] [W] [TRT] onnx2trt_utils.cpp:375: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
+[01/04/2024-15:28:15] [W] [TRT] onnx2trt_utils.cpp:403: One or more weights outside the range of INT32 was clamped
+[01/04/2024-15:28:18] [W] [TRT] Calibrator won't be used in explicit precision mode. Use quantization aware training to generate network with Quantize/Dequantize nodes.
+[01/04/2024-15:38:46] [W] * Throughput may be bound by Enqueue Time rather than GPU Compute and the GPU may be under-utilized.
+[01/04/2024-15:38:46] [W]   If not already in use, --useCudaGraph (utilize CUDA graphs where possible) may increase the throughput.
+[01/04/2024-15:38:46] [W] * GPU compute time is unstable, with coefficient of variance = 4.08535%.
+[01/04/2024-15:38:46] [W]   If not already in use, locking GPU clock frequency or adding --useSpinWait may improve the stability.

yolo_nas_pose_s_int8.onnx.int8.engine.log

@@ -1,322 +1,324 @@
 &&&& RUNNING TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_s_int8.onnx --int8 --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_s_int8.onnx.int8.engine
-[12/28/2023-03:36:14] [I] === Model Options ===
-[12/28/2023-03:36:14] [I] Format: ONNX
-[12/28/2023-03:36:14] [I] Model: yolo_nas_pose_s_int8.onnx
-[12/28/2023-03:36:14] [I] Output:
-[12/28/2023-03:36:14] [I] === Build Options ===
-[12/28/2023-03:36:14] [I] Max batch: explicit batch
-[12/28/2023-03:36:14] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
-[12/28/2023-03:36:14] [I] minTiming: 1
-[12/28/2023-03:36:14] [I] avgTiming: 8
-[12/28/2023-03:36:14] [I] Precision: FP32+INT8
-[12/28/2023-03:36:14] [I] LayerPrecisions: 
-[12/28/2023-03:36:14] [I] Calibration: Dynamic
-[12/28/2023-03:36:14] [I] Refit: Disabled
-[12/28/2023-03:36:14] [I] Sparsity: Disabled
-[12/28/2023-03:36:14] [I] Safe mode: Disabled
-[12/28/2023-03:36:14] [I] DirectIO mode: Disabled
-[12/28/2023-03:36:14] [I] Restricted mode: Disabled
-[12/28/2023-03:36:14] [I] Build only: Disabled
-[12/28/2023-03:36:14] [I] Save engine: yolo_nas_pose_s_int8.onnx.int8.engine
-[12/28/2023-03:36:14] [I] Load engine: 
-[12/28/2023-03:36:14] [I] Profiling verbosity: 0
-[12/28/2023-03:36:14] [I] Tactic sources: Using default tactic sources
-[12/28/2023-03:36:14] [I] timingCacheMode: local
-[12/28/2023-03:36:14] [I] timingCacheFile: 
-[12/28/2023-03:36:14] [I] Heuristic: Disabled
-[12/28/2023-03:36:14] [I] Preview Features: Use default preview flags.
-[12/28/2023-03:36:14] [I] Input(s)s format: fp32:CHW
-[12/28/2023-03:36:14] [I] Output(s)s format: fp32:CHW
-[12/28/2023-03:36:14] [I] Input build shapes: model
-[12/28/2023-03:36:14] [I] Input calibration shapes: model
-[12/28/2023-03:36:14] [I] === System Options ===
-[12/28/2023-03:36:14] [I] Device: 0
-[12/28/2023-03:36:14] [I] DLACore: 
-[12/28/2023-03:36:14] [I] Plugins:
-[12/28/2023-03:36:14] [I] === Inference Options ===
-[12/28/2023-03:36:14] [I] Batch: Explicit
-[12/28/2023-03:36:14] [I] Input inference shapes: model
-[12/28/2023-03:36:14] [I] Iterations: 10
-[12/28/2023-03:36:14] [I] Duration: 15s (+ 200ms warm up)
-[12/28/2023-03:36:14] [I] Sleep time: 0ms
-[12/28/2023-03:36:14] [I] Idle time: 0ms
-[12/28/2023-03:36:14] [I] Streams: 1
-[12/28/2023-03:36:14] [I] ExposeDMA: Disabled
-[12/28/2023-03:36:14] [I] Data transfers: Enabled
-[12/28/2023-03:36:14] [I] Spin-wait: Disabled
-[12/28/2023-03:36:14] [I] Multithreading: Disabled
-[12/28/2023-03:36:14] [I] CUDA Graph: Disabled
-[12/28/2023-03:36:14] [I] Separate profiling: Disabled
-[12/28/2023-03:36:14] [I] Time Deserialize: Disabled
-[12/28/2023-03:36:14] [I] Time Refit: Disabled
-[12/28/2023-03:36:14] [I] NVTX verbosity: 0
-[12/28/2023-03:36:14] [I] Persistent Cache Ratio: 0
-[12/28/2023-03:36:14] [I] Inputs:
-[12/28/2023-03:36:14] [I] === Reporting Options ===
-[12/28/2023-03:36:14] [I] Verbose: Disabled
-[12/28/2023-03:36:14] [I] Averages: 100 inferences
-[12/28/2023-03:36:14] [I] Percentiles: 90,95,99
-[12/28/2023-03:36:14] [I] Dump refittable layers:Disabled
-[12/28/2023-03:36:14] [I] Dump output: Disabled
-[12/28/2023-03:36:14] [I] Profile: Disabled
-[12/28/2023-03:36:14] [I] Export timing to JSON file: 
-[12/28/2023-03:36:14] [I] Export output to JSON file: 
-[12/28/2023-03:36:14] [I] Export profile to JSON file: 
-[12/28/2023-03:36:14] [I] 
-[12/28/2023-03:36:14] [I] === Device Information ===
-[12/28/2023-03:36:14] [I] Selected Device: Orin
-[12/28/2023-03:36:14] [I] Compute Capability: 8.7
-[12/28/2023-03:36:14] [I] SMs: 8
-[12/28/2023-03:36:14] [I] Compute Clock Rate: 0.624 GHz
-[12/28/2023-03:36:14] [I] Device Global Memory: 7471 MiB
-[12/28/2023-03:36:14] [I] Shared Memory per SM: 164 KiB
-[12/28/2023-03:36:14] [I] Memory Bus Width: 128 bits (ECC disabled)
-[12/28/2023-03:36:14] [I] Memory Clock Rate: 0.624 GHz
-[12/28/2023-03:36:14] [I] 
-[12/28/2023-03:36:14] [I] TensorRT version: 8.5.2
-[12/28/2023-03:36:15] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 2931 (MiB)
-[12/28/2023-03:36:17] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +285, now: CPU 574, GPU 3236 (MiB)
-[12/28/2023-03:36:18] [I] Start parsing network model
-[12/28/2023-03:36:18] [I] [TRT] ----------------------------------------------------------------
-[12/28/2023-03:36:18] [I] [TRT] Input filename:   yolo_nas_pose_s_int8.onnx
-[12/28/2023-03:36:18] [I] [TRT] ONNX IR version:  0.0.8
-[12/28/2023-03:36:18] [I] [TRT] Opset version:    17
-[12/28/2023-03:36:18] [I] [TRT] Producer name:    pytorch
-[12/28/2023-03:36:18] [I] [TRT] Producer version: 2.1.2
-[12/28/2023-03:36:18] [I] [TRT] Domain:           
-[12/28/2023-03:36:18] [I] [TRT] Model version:    0
-[12/28/2023-03:36:18] [I] [TRT] Doc string:       
-[12/28/2023-03:36:18] [I] [TRT] ----------------------------------------------------------------
-[12/28/2023-03:36:21] [I] Finish parsing network model
-[12/28/2023-03:36:21] [I] FP32 and INT8 precisions have been specified - more performance might be enabled by additionally specifying --fp16 or --best
-[12/28/2023-03:36:24] [I] [TRT] ---------- Layers Running on DLA ----------
-[12/28/2023-03:36:24] [I] [TRT] ---------- Layers Running on GPU ----------
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1228) [Constant]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1229) [Constant]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1230) [Constant]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 494) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 510) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 557) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 573) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 589) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 639) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 655) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 671) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 687) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.4.alpha + (Unnamed Layer* 703) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 744) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 760) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 825) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 841) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 890) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 906) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/Concat_/model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 972) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1013) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv || model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1078) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1119) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] NMS: batched_nms_243
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1232) [NMS]_1_output[DevicetoShapeHostCopy]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
-[12/28/2023-03:36:24] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
-[12/28/2023-03:36:26] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +773, now: CPU 1233, GPU 4134 (MiB)
-[12/28/2023-03:36:26] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +118, now: CPU 1315, GPU 4252 (MiB)
-[12/28/2023-03:36:26] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
-[12/28/2023-03:46:28] [I] [TRT] Total Activation Memory: 7940674048
-[12/28/2023-03:46:28] [I] [TRT] Detected 1 inputs and 1 output network tensors.
-[12/28/2023-03:46:29] [I] [TRT] Total Host Persistent Memory: 307232
-[12/28/2023-03:46:29] [I] [TRT] Total Device Persistent Memory: 38912
-[12/28/2023-03:46:29] [I] [TRT] Total Scratch Memory: 134217728
-[12/28/2023-03:46:29] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 74 MiB, GPU 132 MiB
-[12/28/2023-03:46:29] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 173 steps to complete.
-[12/28/2023-03:46:29] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 42.4134ms to assign 13 blocks to 173 nodes requiring 144692224 bytes.
-[12/28/2023-03:46:29] [I] [TRT] Total Activation Memory: 144692224
-[12/28/2023-03:46:30] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +6, now: CPU 1665, GPU 5446 (MiB)
-[12/28/2023-03:46:30] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +15, GPU +16, now: CPU 15, GPU 16 (MiB)
-[12/28/2023-03:46:31] [I] Engine built in 616.071 sec.
-[12/28/2023-03:46:31] [I] [TRT] Loaded engine size: 17 MiB
-[12/28/2023-03:46:31] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +12, now: CPU 1250, GPU 5268 (MiB)
-[12/28/2023-03:46:31] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +15, now: CPU 0, GPU 15 (MiB)
-[12/28/2023-03:46:31] [I] Engine deserialized in 0.135656 sec.
-[12/28/2023-03:46:31] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +7, now: CPU 1251, GPU 5268 (MiB)
-[12/28/2023-03:46:31] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +138, now: CPU 0, GPU 153 (MiB)
-[12/28/2023-03:46:31] [I] Setting persistentCacheLimit to 0 bytes.
-[12/28/2023-03:46:31] [I] Using random values for input onnx::Cast_0
-[12/28/2023-03:46:31] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
-[12/28/2023-03:46:31] [I] Using random values for output graph2_flat_predictions
-[12/28/2023-03:46:31] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
-[12/28/2023-03:46:31] [I] Starting inference
-[12/28/2023-03:46:46] [I] Warmup completed 8 queries over 200 ms
-[12/28/2023-03:46:46] [I] Timing trace has 959 queries over 15.0421 s
-[12/28/2023-03:46:46] [I] 
-[12/28/2023-03:46:46] [I] === Trace details ===
-[12/28/2023-03:46:46] [I] Trace averages of 100 runs:
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.6862 ms - Host latency: 15.799 ms (enqueue 15.7577 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.8638 ms - Host latency: 15.9795 ms (enqueue 15.9265 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 16.1783 ms - Host latency: 16.2971 ms (enqueue 16.24 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.4097 ms - Host latency: 15.5201 ms (enqueue 15.487 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.3654 ms - Host latency: 15.4758 ms (enqueue 15.4444 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.4303 ms - Host latency: 15.5405 ms (enqueue 15.5048 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.388 ms - Host latency: 15.5003 ms (enqueue 15.463 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.3653 ms - Host latency: 15.4758 ms (enqueue 15.4443 ms)
-[12/28/2023-03:46:46] [I] Average on 100 runs - GPU latency: 15.3726 ms - Host latency: 15.4828 ms (enqueue 15.4505 ms)
-[12/28/2023-03:46:46] [I] 
-[12/28/2023-03:46:46] [I] === Performance summary ===
-[12/28/2023-03:46:46] [I] Throughput: 63.7546 qps
-[12/28/2023-03:46:46] [I] Latency: min = 14.4111 ms, max = 20.8093 ms, mean = 15.6545 ms, median = 15.4062 ms, percentile(90%) = 16.3127 ms, percentile(95%) = 16.4324 ms, percentile(99%) = 17.4062 ms
-[12/28/2023-03:46:46] [I] Enqueue Time: min = 14.3828 ms, max = 20.7617 ms, mean = 15.6158 ms, median = 15.3755 ms, percentile(90%) = 16.2577 ms, percentile(95%) = 16.3701 ms, percentile(99%) = 17.3357 ms
-[12/28/2023-03:46:46] [I] H2D Latency: min = 0.0805664 ms, max = 0.117188 ms, mean = 0.0972066 ms, median = 0.0976562 ms, percentile(90%) = 0.0996094 ms, percentile(95%) = 0.0998535 ms, percentile(99%) = 0.111328 ms
-[12/28/2023-03:46:46] [I] GPU Compute Time: min = 14.3018 ms, max = 20.6875 ms, mean = 15.5423 ms, median = 15.2939 ms, percentile(90%) = 16.1956 ms, percentile(95%) = 16.3105 ms, percentile(99%) = 17.293 ms
-[12/28/2023-03:46:46] [I] D2H Latency: min = 0.00292969 ms, max = 0.0727539 ms, mean = 0.014973 ms, median = 0.0120239 ms, percentile(90%) = 0.0234375 ms, percentile(95%) = 0.0253906 ms, percentile(99%) = 0.0373535 ms
-[12/28/2023-03:46:46] [I] Total Host Walltime: 15.0421 s
-[12/28/2023-03:46:46] [I] Total GPU Compute Time: 14.9051 s
-[12/28/2023-03:46:46] [I] Explanations of the performance metrics are printed in the verbose logs.
-[12/28/2023-03:46:46] [I] 
+[01/04/2024-15:28:11] [I] === Model Options ===
+[01/04/2024-15:28:11] [I] Format: ONNX
+[01/04/2024-15:28:11] [I] Model: yolo_nas_pose_s_int8.onnx
+[01/04/2024-15:28:11] [I] Output:
+[01/04/2024-15:28:11] [I] === Build Options ===
+[01/04/2024-15:28:11] [I] Max batch: explicit batch
+[01/04/2024-15:28:11] [I] Memory Pools: workspace: default, dlaSRAM: default, dlaLocalDRAM: default, dlaGlobalDRAM: default
+[01/04/2024-15:28:11] [I] minTiming: 1
+[01/04/2024-15:28:11] [I] avgTiming: 8
+[01/04/2024-15:28:11] [I] Precision: FP32+INT8
+[01/04/2024-15:28:11] [I] LayerPrecisions: 
+[01/04/2024-15:28:11] [I] Calibration: Dynamic
+[01/04/2024-15:28:11] [I] Refit: Disabled
+[01/04/2024-15:28:11] [I] Sparsity: Disabled
+[01/04/2024-15:28:11] [I] Safe mode: Disabled
+[01/04/2024-15:28:11] [I] DirectIO mode: Disabled
+[01/04/2024-15:28:11] [I] Restricted mode: Disabled
+[01/04/2024-15:28:11] [I] Build only: Disabled
+[01/04/2024-15:28:11] [I] Save engine: yolo_nas_pose_s_int8.onnx.int8.engine
+[01/04/2024-15:28:11] [I] Load engine: 
+[01/04/2024-15:28:11] [I] Profiling verbosity: 0
+[01/04/2024-15:28:11] [I] Tactic sources: Using default tactic sources
+[01/04/2024-15:28:11] [I] timingCacheMode: local
+[01/04/2024-15:28:11] [I] timingCacheFile: 
+[01/04/2024-15:28:11] [I] Heuristic: Disabled
+[01/04/2024-15:28:11] [I] Preview Features: Use default preview flags.
+[01/04/2024-15:28:11] [I] Input(s)s format: fp32:CHW
+[01/04/2024-15:28:11] [I] Output(s)s format: fp32:CHW
+[01/04/2024-15:28:11] [I] Input build shapes: model
+[01/04/2024-15:28:11] [I] Input calibration shapes: model
+[01/04/2024-15:28:11] [I] === System Options ===
+[01/04/2024-15:28:11] [I] Device: 0
+[01/04/2024-15:28:11] [I] DLACore: 
+[01/04/2024-15:28:11] [I] Plugins:
+[01/04/2024-15:28:11] [I] === Inference Options ===
+[01/04/2024-15:28:11] [I] Batch: Explicit
+[01/04/2024-15:28:11] [I] Input inference shapes: model
+[01/04/2024-15:28:11] [I] Iterations: 10
+[01/04/2024-15:28:11] [I] Duration: 15s (+ 200ms warm up)
+[01/04/2024-15:28:11] [I] Sleep time: 0ms
+[01/04/2024-15:28:11] [I] Idle time: 0ms
+[01/04/2024-15:28:11] [I] Streams: 1
+[01/04/2024-15:28:11] [I] ExposeDMA: Disabled
+[01/04/2024-15:28:11] [I] Data transfers: Enabled
+[01/04/2024-15:28:11] [I] Spin-wait: Disabled
+[01/04/2024-15:28:11] [I] Multithreading: Disabled
+[01/04/2024-15:28:11] [I] CUDA Graph: Disabled
+[01/04/2024-15:28:11] [I] Separate profiling: Disabled
+[01/04/2024-15:28:11] [I] Time Deserialize: Disabled
+[01/04/2024-15:28:11] [I] Time Refit: Disabled
+[01/04/2024-15:28:11] [I] NVTX verbosity: 0
+[01/04/2024-15:28:11] [I] Persistent Cache Ratio: 0
+[01/04/2024-15:28:11] [I] Inputs:
+[01/04/2024-15:28:11] [I] === Reporting Options ===
+[01/04/2024-15:28:11] [I] Verbose: Disabled
+[01/04/2024-15:28:11] [I] Averages: 100 inferences
+[01/04/2024-15:28:11] [I] Percentiles: 90,95,99
+[01/04/2024-15:28:11] [I] Dump refittable layers:Disabled
+[01/04/2024-15:28:11] [I] Dump output: Disabled
+[01/04/2024-15:28:11] [I] Profile: Disabled
+[01/04/2024-15:28:11] [I] Export timing to JSON file: 
+[01/04/2024-15:28:11] [I] Export output to JSON file: 
+[01/04/2024-15:28:11] [I] Export profile to JSON file: 
+[01/04/2024-15:28:11] [I] 
+[01/04/2024-15:28:11] [I] === Device Information ===
+[01/04/2024-15:28:11] [I] Selected Device: Orin
+[01/04/2024-15:28:11] [I] Compute Capability: 8.7
+[01/04/2024-15:28:11] [I] SMs: 8
+[01/04/2024-15:28:11] [I] Compute Clock Rate: 0.624 GHz
+[01/04/2024-15:28:11] [I] Device Global Memory: 7471 MiB
+[01/04/2024-15:28:11] [I] Shared Memory per SM: 164 KiB
+[01/04/2024-15:28:11] [I] Memory Bus Width: 128 bits (ECC disabled)
+[01/04/2024-15:28:11] [I] Memory Clock Rate: 0.624 GHz
+[01/04/2024-15:28:11] [I] 
+[01/04/2024-15:28:11] [I] TensorRT version: 8.5.2
+[01/04/2024-15:28:11] [I] [TRT] [MemUsageChange] Init CUDA: CPU +220, GPU +0, now: CPU 249, GPU 3760 (MiB)
+[01/04/2024-15:28:15] [I] [TRT] [MemUsageChange] Init builder kernel library: CPU +302, GPU +284, now: CPU 574, GPU 4064 (MiB)
+[01/04/2024-15:28:15] [I] Start parsing network model
+[01/04/2024-15:28:15] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-15:28:15] [I] [TRT] Input filename:   yolo_nas_pose_s_int8.onnx
+[01/04/2024-15:28:15] [I] [TRT] ONNX IR version:  0.0.8
+[01/04/2024-15:28:15] [I] [TRT] Opset version:    17
+[01/04/2024-15:28:15] [I] [TRT] Producer name:    pytorch
+[01/04/2024-15:28:15] [I] [TRT] Producer version: 2.1.2
+[01/04/2024-15:28:15] [I] [TRT] Domain:           
+[01/04/2024-15:28:15] [I] [TRT] Model version:    0
+[01/04/2024-15:28:15] [I] [TRT] Doc string:       
+[01/04/2024-15:28:15] [I] [TRT] ----------------------------------------------------------------
+[01/04/2024-15:28:18] [I] Finish parsing network model
+[01/04/2024-15:28:18] [I] FP32 and INT8 precisions have been specified - more performance might be enabled by additionally specifying --fp16 or --best
+[01/04/2024-15:28:22] [I] [TRT] ---------- Layers Running on DLA ----------
+[01/04/2024-15:28:22] [I] [TRT] ---------- Layers Running on GPU ----------
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation1]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/pre_process/pre_process.0/Cast.../pre_process/pre_process.2/Mul]}
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1229) [Constant]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1230) [Constant]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONSTANT: (Unnamed Layer* 1231) [Constant]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stem/conv/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stem.conv.rbr_reparam.weight + /model/backbone/stem/conv/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stem/conv/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.downsample.rbr_reparam.weight + /model/backbone/stage1/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/downsample/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv2.conv.weight + /model/backbone/stage1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv1.conv.weight + /model/backbone/stage1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 494) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 510) [Shuffle] + /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage1.blocks.conv3.conv.weight + /model/backbone/stage1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage1/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_skip2.conv.weight + /model/neck/neck2/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.downsample.rbr_reparam.weight + /model/backbone/stage2/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/downsample/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.downsample.conv.weight + /model/neck/neck2/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/downsample/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv2.conv.weight + /model/backbone/stage2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv1.conv.weight + /model/backbone/stage2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 557) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 573) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage2.blocks.bottlenecks.2.alpha + (Unnamed Layer* 589) [Shuffle] + /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage2/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage2.blocks.conv3.conv.weight + /model/backbone/stage2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage2/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip2.conv.weight + /model/neck/neck1/reduce_skip2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip2/conv/Conv || model.neck.neck2.reduce_skip1.conv.weight + /model/neck/neck2/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_skip1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.downsample.rbr_reparam.weight + /model/backbone/stage3/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/downsample/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.downsample.conv.weight + /model/neck/neck1/downsample/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/downsample/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv2.conv.weight + /model/backbone/stage3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv1.conv.weight + /model/backbone/stage3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 639) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 655) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.2.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.2.alpha + (Unnamed Layer* 671) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.2/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.3.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.3.alpha + (Unnamed Layer* 687) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.3/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv1.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.bottlenecks.4.cv2.rbr_reparam.weight + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage3.blocks.bottlenecks.4.alpha + (Unnamed Layer* 703) [Shuffle] + /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Mul, /model/backbone/stage3/blocks/bottlenecks/bottlenecks.4/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage3.blocks.conv3.conv.weight + /model/backbone/stage3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage3/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_skip1.conv.weight + /model/neck/neck1/reduce_skip1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_skip1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.downsample.rbr_reparam.weight + /model/backbone/stage4/downsample/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/downsample/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv2.conv.weight + /model/backbone/stage4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv1.conv.weight + /model/backbone/stage4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 744) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.backbone.stage4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 760) [Shuffle] + /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Mul, /model/backbone/stage4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.stage4.blocks.conv3.conv.weight + /model/backbone/stage4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/backbone/stage4/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv1.conv.weight + /model/backbone/context_module/cv1/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.2/MaxPool
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.1/MaxPool
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POOLING: /model/backbone/context_module/m.0/MaxPool
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/backbone/context_module/m.2/MaxPool_output_0 copy
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.backbone.context_module.cv2.conv.weight + /model/backbone/context_module/cv2/conv/_weight_quantizer/QuantizeLinear + /model/backbone/context_module/cv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.conv.conv.weight + /model/neck/neck1/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/conv/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck1.upsample.weight + /model/neck/neck1/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck1/upsample/ConvTranspose
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.reduce_after_concat.conv.weight + /model/neck/neck1/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/reduce_after_concat/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv2.conv.weight + /model/neck/neck1/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv1.conv.weight + /model/neck/neck1/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.0.alpha + (Unnamed Layer* 825) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck1.blocks.bottlenecks.1.alpha + (Unnamed Layer* 841) [Shuffle] + /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck1/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck1.blocks.conv3.conv.weight + /model/neck/neck1/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck1/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.conv.conv.weight + /model/neck/neck2/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/conv/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/upsample/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] DECONVOLUTION: model.neck.neck2.upsample.weight + /model/neck/neck2/upsample/_weight_quantizer/QuantizeLinear + /model/neck/neck2/upsample/ConvTranspose
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/Concat_/model/neck/neck2/reduce_skip1/act/Relu_output_0_clone_1 copy
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.reduce_after_concat.conv.weight + /model/neck/neck2/reduce_after_concat/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/reduce_after_concat/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv2.conv.weight + /model/neck/neck2/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv1.conv.weight + /model/neck/neck2/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.0.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.0.alpha + (Unnamed Layer* 890) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv1.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv1/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.bottlenecks.1.cv2.rbr_reparam.weight + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/cv2/rbr_reparam/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck2.blocks.bottlenecks.1.alpha + (Unnamed Layer* 906) [Shuffle] + /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck2/blocks/Concat_/model/neck/neck2/blocks/bottlenecks/bottlenecks.1/Add_output_0_clone_0 copy
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck2.blocks.conv3.conv.weight + /model/neck/neck2/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck2/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.bbox_stem.seq.conv.weight + /model/heads/head1/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/bbox_stem/seq/conv/Conv || model.heads.head1.pose_stem.seq.conv.weight + /model/heads/head1/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_stem/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.conv.conv.weight + /model/neck/neck3/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/conv/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_convs.0.seq.conv.weight + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head1.cls_convs.0.seq.conv.weight + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.0.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.cls_pred.weight + /model/heads/head1/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/cls_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.reg_pred.weight + /model/heads/head1/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/reg_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_convs.1.seq.conv.weight + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv2.conv.weight + /model/neck/neck3/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv1.conv.weight + /model/neck/neck3/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape + /model/heads/Transpose
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head1.pose_pred.weight + /model/heads/head1/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head1/pose_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.0.alpha + (Unnamed Layer* 972) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck3.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1013) [Shuffle] + /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck3/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck3.blocks.conv3.conv.weight + /model/neck/neck3/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck3/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_stem.seq.conv.weight + /model/heads/head2/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_stem/seq/conv/Conv || model.heads.head2.bbox_stem.seq.conv.weight + /model/heads/head2/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/bbox_stem/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.conv.conv.weight + /model/neck/neck4/conv/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/conv/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_convs.0.seq.conv.weight + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head2.cls_convs.0.seq.conv.weight + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.0.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/conv1/conv/_input_quantizer/QuantizeLinear_clone_1
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.cls_pred.weight + /model/heads/head2/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/cls_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.reg_pred.weight + /model/heads/head2/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/reg_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_convs.1.seq.conv.weight + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv2.conv.weight + /model/neck/neck4/blocks/conv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv1.conv.weight + /model/neck/neck4/blocks/conv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_4 + /model/heads/Transpose_3
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head2.pose_pred.weight + /model/heads/head2/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head2/pose_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_1
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_1
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.0.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/cv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.0.alpha + (Unnamed Layer* 1079) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.0/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] COPY: /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_input_quantizer/QuantizeLinear
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv1.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv1/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.bottlenecks.1.cv2.conv.weight + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/cv2/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] POINTWISE: PWN(model.neck.neck4.blocks.bottlenecks.1.alpha + (Unnamed Layer* 1120) [Shuffle] + /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Mul, /model/neck/neck4/blocks/bottlenecks/bottlenecks.1/Add)
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.neck.neck4.blocks.conv3.conv.weight + /model/neck/neck4/blocks/conv3/conv/_weight_quantizer/QuantizeLinear + /model/neck/neck4/blocks/conv3/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.bbox_stem.seq.conv.weight + /model/heads/head3/bbox_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/bbox_stem/seq/conv/Conv || model.heads.head3.pose_stem.seq.conv.weight + /model/heads/head3/pose_stem/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_stem/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_convs.0.seq.conv.weight + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_convs/reg_convs.0/seq/conv/Conv || model.heads.head3.cls_convs.0.seq.conv.weight + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_convs/cls_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.0.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.0/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.cls_pred.weight + /model/heads/head3/cls_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/cls_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.reg_pred.weight + /model/heads/head3/reg_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/reg_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.1.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.1/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SHUFFLE: /model/heads/Reshape_8 + /model/heads/Transpose_6
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_convs.2.seq.conv.weight + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_convs/pose_convs.2/seq/conv/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] SOFTMAX: /model/heads/Softmax_2
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: model.heads.head3.pose_pred.weight + /model/heads/head3/pose_pred/_weight_quantizer/QuantizeLinear + /model/heads/head3/pose_pred/Conv
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] CONVOLUTION: /model/heads/Conv_2
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice_1.../post_process/Reshape_2]}
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] NMS: batched_nms_243
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] DEVICE_TO_SHAPE_HOST: (Unnamed Layer* 1233) [NMS]_1_output[DevicetoShapeHostCopy]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation2]
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] MYELIN: {ForeignNode[/model/heads/head1/Slice...graph2_/Concat_5]}
+[01/04/2024-15:28:22] [I] [TRT] [GpuLayer] TRAIN_STATION: [trainStation3]
+[01/04/2024-15:28:23] [I] [TRT] [MemUsageChange] Init cuBLAS/cuBLASLt: CPU +534, GPU +499, now: CPU 1233, GPU 4691 (MiB)
+[01/04/2024-15:28:23] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +82, GPU +74, now: CPU 1315, GPU 4765 (MiB)
+[01/04/2024-15:28:23] [I] [TRT] Local timing cache in use. Profiling results in this builder pass will not be stored.
+[01/04/2024-15:38:28] [I] [TRT] Total Activation Memory: 7939057152
+[01/04/2024-15:38:28] [I] [TRT] Detected 1 inputs and 1 output network tensors.
+[01/04/2024-15:38:29] [I] [TRT] Total Host Persistent Memory: 309280
+[01/04/2024-15:38:29] [I] [TRT] Total Device Persistent Memory: 38912
+[01/04/2024-15:38:29] [I] [TRT] Total Scratch Memory: 134217728
+[01/04/2024-15:38:29] [I] [TRT] [MemUsageStats] Peak memory usage of TRT CPU/GPU memory allocators: CPU 74 MiB, GPU 132 MiB
+[01/04/2024-15:38:29] [I] [TRT] [BlockAssignment] Started assigning block shifts. This will take 171 steps to complete.
+[01/04/2024-15:38:29] [I] [TRT] [BlockAssignment] Algorithm ShiftNTopDown took 40.1195ms to assign 13 blocks to 171 nodes requiring 144747520 bytes.
+[01/04/2024-15:38:29] [I] [TRT] Total Activation Memory: 144747520
+[01/04/2024-15:38:30] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1665, GPU 5901 (MiB)
+[01/04/2024-15:38:30] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in building engine: CPU +15, GPU +16, now: CPU 15, GPU 16 (MiB)
+[01/04/2024-15:38:30] [I] Engine built in 619.407 sec.
+[01/04/2024-15:38:31] [I] [TRT] Loaded engine size: 17 MiB
+[01/04/2024-15:38:31] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +1, GPU +0, now: CPU 1251, GPU 5877 (MiB)
+[01/04/2024-15:38:31] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +15, now: CPU 0, GPU 15 (MiB)
+[01/04/2024-15:38:31] [I] Engine deserialized in 0.128553 sec.
+[01/04/2024-15:38:31] [I] [TRT] [MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 1251, GPU 5877 (MiB)
+[01/04/2024-15:38:31] [I] [TRT] [MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +138, now: CPU 0, GPU 153 (MiB)
+[01/04/2024-15:38:31] [I] Setting persistentCacheLimit to 0 bytes.
+[01/04/2024-15:38:31] [I] Using random values for input onnx::Cast_0
+[01/04/2024-15:38:31] [I] Created input binding for onnx::Cast_0 with dimensions 1x3x640x640
+[01/04/2024-15:38:31] [I] Using random values for output graph2_flat_predictions
+[01/04/2024-15:38:31] [I] Created output binding for graph2_flat_predictions with dimensions -1x57
+[01/04/2024-15:38:31] [I] Starting inference
+[01/04/2024-15:38:46] [I] Warmup completed 10 queries over 200 ms
+[01/04/2024-15:38:46] [I] Timing trace has 924 queries over 15.0277 s
+[01/04/2024-15:38:46] [I] 
+[01/04/2024-15:38:46] [I] === Trace details ===
+[01/04/2024-15:38:46] [I] Trace averages of 100 runs:
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 16.5012 ms - Host latency: 16.6184 ms (enqueue 16.5675 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 16.52 ms - Host latency: 16.6389 ms (enqueue 16.5841 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 15.9968 ms - Host latency: 16.11 ms (enqueue 16.0689 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 15.7859 ms - Host latency: 15.8982 ms (enqueue 15.8629 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 15.6698 ms - Host latency: 15.7823 ms (enqueue 15.7438 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 16.1195 ms - Host latency: 16.2344 ms (enqueue 16.191 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 16.3087 ms - Host latency: 16.4258 ms (enqueue 16.3729 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 15.5609 ms - Host latency: 15.6727 ms (enqueue 15.6373 ms)
+[01/04/2024-15:38:46] [I] Average on 100 runs - GPU latency: 16.4593 ms - Host latency: 16.577 ms (enqueue 16.5201 ms)
+[01/04/2024-15:38:46] [I] 
+[01/04/2024-15:38:46] [I] === Performance summary ===
+[01/04/2024-15:38:46] [I] Throughput: 61.4865 qps
+[01/04/2024-15:38:46] [I] Latency: min = 14.6812 ms, max = 18.0088 ms, mean = 16.2285 ms, median = 16.2705 ms, percentile(90%) = 17.0212 ms, percentile(95%) = 17.188 ms, percentile(99%) = 17.4453 ms
+[01/04/2024-15:38:46] [I] Enqueue Time: min = 14.6509 ms, max = 17.9592 ms, mean = 16.1828 ms, median = 16.2139 ms, percentile(90%) = 16.9602 ms, percentile(95%) = 17.1279 ms, percentile(99%) = 17.3662 ms
+[01/04/2024-15:38:46] [I] H2D Latency: min = 0.0830078 ms, max = 0.121094 ms, mean = 0.0968298 ms, median = 0.0967102 ms, percentile(90%) = 0.0991211 ms, percentile(95%) = 0.0996094 ms, percentile(99%) = 0.112915 ms
+[01/04/2024-15:38:46] [I] GPU Compute Time: min = 14.5688 ms, max = 17.8901 ms, mean = 16.1133 ms, median = 16.1515 ms, percentile(90%) = 16.9043 ms, percentile(95%) = 17.0674 ms, percentile(99%) = 17.3271 ms
+[01/04/2024-15:38:46] [I] D2H Latency: min = 0.00341797 ms, max = 0.0561523 ms, mean = 0.0183549 ms, median = 0.0194397 ms, percentile(90%) = 0.0244141 ms, percentile(95%) = 0.0263672 ms, percentile(99%) = 0.0390625 ms
+[01/04/2024-15:38:46] [I] Total Host Walltime: 15.0277 s
+[01/04/2024-15:38:46] [I] Total GPU Compute Time: 14.8887 s
+[01/04/2024-15:38:46] [I] Explanations of the performance metrics are printed in the verbose logs.
+[01/04/2024-15:38:46] [I] 
 &&&& PASSED TensorRT.trtexec [TensorRT v8502] # /usr/src/tensorrt/bin/trtexec --onnx=yolo_nas_pose_s_int8.onnx --int8 --avgRuns=100 --duration=15 --saveEngine=yolo_nas_pose_s_int8.onnx.int8.engine