Training in progress, step 3000, checkpoint

checkpoint-3000/config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "openai/clip-vit-large-patch14-336",
+  "architectures": [
+    "CLIPModel"
+  ],
+  "initializer_factor": 1.0,
+  "logit_scale_init_value": 2.6592,
+  "model_type": "clip",
+  "projection_dim": 768,
+  "text_config": {
+    "dropout": 0.0,
+    "hidden_size": 768,
+    "intermediate_size": 3072,
+    "model_type": "clip_text_model",
+    "num_attention_heads": 12,
+    "projection_dim": 768
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.45.0.dev0",
+  "vision_config": {
+    "dropout": 0.0,
+    "hidden_size": 1024,
+    "image_size": 336,
+    "intermediate_size": 4096,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768
+  }
+}

checkpoint-3000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33246f47d2c738fc4b6e1f2b15e83f9f22ff7f422b2e4444d177aaef1d6b19b5
+size 1711848436

checkpoint-3000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f74c67a84f0518ccaa360b88de1c042646a489af734d4467bbc7933a64764776
+size 3424043887

checkpoint-3000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9733db299f85727654ada1e2241c2b4f64948c1fd868b50f2037c2a0349da2
+size 14503

checkpoint-3000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bd741cfbf67d8dd3aa2ffc45f4be323c445e2a4df6aa6318a1d97a5e185cee9d
+size 623

checkpoint-3000/trainer_state.json

@@ -0,0 +1,123 @@
+{
+  "best_metric": 2.0204577445983887,
+  "best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e3l55-l/checkpoint-3000",
+  "epoch": 0.5525879535826119,
+  "eval_steps": 500,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09209799226376865,
+      "grad_norm": 0.009597355499863625,
+      "learning_rate": 4.846503346227052e-05,
+      "loss": 0.7336,
+      "step": 500
+    },
+    {
+      "epoch": 0.09209799226376865,
+      "eval_loss": 2.0780556201934814,
+      "eval_runtime": 77.1404,
+      "eval_samples_per_second": 15.634,
+      "eval_steps_per_second": 1.957,
+      "step": 500
+    },
+    {
+      "epoch": 0.1841959845275373,
+      "grad_norm": 0.10049372166395187,
+      "learning_rate": 4.693006692454105e-05,
+      "loss": 0.6953,
+      "step": 1000
+    },
+    {
+      "epoch": 0.1841959845275373,
+      "eval_loss": 2.0782744884490967,
+      "eval_runtime": 77.8965,
+      "eval_samples_per_second": 15.482,
+      "eval_steps_per_second": 1.938,
+      "step": 1000
+    },
+    {
+      "epoch": 0.27629397679130596,
+      "grad_norm": 0.0041180383414030075,
+      "learning_rate": 4.539510038681157e-05,
+      "loss": 0.6953,
+      "step": 1500
+    },
+    {
+      "epoch": 0.27629397679130596,
+      "eval_loss": 2.077699899673462,
+      "eval_runtime": 78.9586,
+      "eval_samples_per_second": 15.274,
+      "eval_steps_per_second": 1.912,
+      "step": 1500
+    },
+    {
+      "epoch": 0.3683919690550746,
+      "grad_norm": 0.29397013783454895,
+      "learning_rate": 4.386013384908209e-05,
+      "loss": 0.6932,
+      "step": 2000
+    },
+    {
+      "epoch": 0.3683919690550746,
+      "eval_loss": 2.0721774101257324,
+      "eval_runtime": 77.349,
+      "eval_samples_per_second": 15.592,
+      "eval_steps_per_second": 1.952,
+      "step": 2000
+    },
+    {
+      "epoch": 0.46048996131884323,
+      "grad_norm": 0.9136261343955994,
+      "learning_rate": 4.232516731135262e-05,
+      "loss": 0.6907,
+      "step": 2500
+    },
+    {
+      "epoch": 0.46048996131884323,
+      "eval_loss": 2.079211711883545,
+      "eval_runtime": 79.3204,
+      "eval_samples_per_second": 15.204,
+      "eval_steps_per_second": 1.904,
+      "step": 2500
+    },
+    {
+      "epoch": 0.5525879535826119,
+      "grad_norm": 4.484213829040527,
+      "learning_rate": 4.079020077362314e-05,
+      "loss": 0.7043,
+      "step": 3000
+    },
+    {
+      "epoch": 0.5525879535826119,
+      "eval_loss": 2.0204577445983887,
+      "eval_runtime": 78.612,
+      "eval_samples_per_second": 15.341,
+      "eval_steps_per_second": 1.921,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 16287,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1079994570482520.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9885fd590ddfbffc4f8db1738c1993002dfa8c1603f7992a1cda9b93bbe7d34
+size 4847