ilsilfverskiold/traffic-levels-image-classification

README.md

@@ -0,0 +1,73 @@
+---
+license: apache-2.0
+base_model: google/vit-base-patch16-224
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+- precision
+- recall
+- f1
+model-index:
+- name: vit-base-patch16-224-finetuned-traffic
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# vit-base-patch16-224-finetuned-traffic
+
+This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4394
+- Accuracy: 0.8292
+- Precision: 0.8232
+- Recall: 0.7366
+- F1: 0.7721
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 128
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 5
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | Precision | Recall | F1     |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|
+| 0.6282        | 0.9843 | 47   | 0.5725          | 0.7644   | 0.7933    | 0.5918 | 0.6525 |
+| 0.4486        | 1.9895 | 95   | 0.4630          | 0.8012   | 0.7964    | 0.6824 | 0.7213 |
+| 0.3285        | 2.9948 | 143  | 0.4394          | 0.8292   | 0.8232    | 0.7366 | 0.7721 |
+| 0.2391        | 4.0    | 191  | 0.4302          | 0.8115   | 0.7941    | 0.7333 | 0.7555 |
+| 0.1814        | 4.9215 | 235  | 0.4365          | 0.8218   | 0.7993    | 0.7362 | 0.7631 |
+
+
+### Framework versions
+
+- Transformers 4.40.1
+- Pytorch 2.2.1+cu121
+- Datasets 2.19.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 4.9214659685863875,
+    "eval_accuracy": 0.8291605301914581,
+    "eval_f1": 0.7721255592297218,
+    "eval_loss": 0.4393855333328247,
+    "eval_precision": 0.823206019483298,
+    "eval_recall": 0.7366059517474952,
+    "eval_runtime": 9.0225,
+    "eval_samples_per_second": 75.256,
+    "eval_steps_per_second": 2.438,
+    "total_flos": 2.328211069271507e+18,
+    "train_loss": 0.4179329159411978,
+    "train_runtime": 724.011,
+    "train_samples_per_second": 42.147,
+    "train_steps_per_second": 0.325
+}

config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "high-traffic",
+    "1": "low-traffic",
+    "2": "medium-traffic",
+    "3": "no-traffic"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "high-traffic": 0,
+    "low-traffic": 1,
+    "medium-traffic": 2,
+    "no-traffic": 3
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1"
+}

eval_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 4.9214659685863875,
+    "eval_accuracy": 0.8291605301914581,
+    "eval_f1": 0.7721255592297218,
+    "eval_loss": 0.4393855333328247,
+    "eval_precision": 0.823206019483298,
+    "eval_recall": 0.7366059517474952,
+    "eval_runtime": 9.0225,
+    "eval_samples_per_second": 75.256,
+    "eval_steps_per_second": 2.438
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3089aefc164aa8950304e18eea0643b15fe7c948950a7c22f596cdead8d4eb5d
+size 343230128

preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

runs/May05_16-51-05_e21ed5085699/events.out.tfevents.1714927879.e21ed5085699.1524.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c68c233bf1cbb5651dfbf8a12ba2e58c31580cab4af475d66412fd0e0dafa15
+size 12361

runs/May05_16-51-05_e21ed5085699/events.out.tfevents.1714928619.e21ed5085699.1524.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:456a53085ec9e94fd28acb50da10ef3783ee6af0c2184b4b549c2342bf07f2b9
+size 560

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.9214659685863875,
+    "total_flos": 2.328211069271507e+18,
+    "train_loss": 0.4179329159411978,
+    "train_runtime": 724.011,
+    "train_samples_per_second": 42.147,
+    "train_steps_per_second": 0.325
+}

trainer_state.json

@@ -0,0 +1,251 @@
+{
+  "best_metric": 0.8291605301914581,
+  "best_model_checkpoint": "vit-base-patch16-224-finetuned-traffic/checkpoint-143",
+  "epoch": 4.9214659685863875,
+  "eval_steps": 500,
+  "global_step": 235,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2094240837696335,
+      "grad_norm": 2.4012229442596436,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 1.2731,
+      "step": 10
+    },
+    {
+      "epoch": 0.418848167539267,
+      "grad_norm": 1.9113624095916748,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.9879,
+      "step": 20
+    },
+    {
+      "epoch": 0.6282722513089005,
+      "grad_norm": 1.7040202617645264,
+      "learning_rate": 4.857819905213271e-05,
+      "loss": 0.762,
+      "step": 30
+    },
+    {
+      "epoch": 0.837696335078534,
+      "grad_norm": 1.6665043830871582,
+      "learning_rate": 4.620853080568721e-05,
+      "loss": 0.6282,
+      "step": 40
+    },
+    {
+      "epoch": 0.9842931937172775,
+      "eval_accuracy": 0.7643593519882179,
+      "eval_f1": 0.6525442813120799,
+      "eval_loss": 0.5724892020225525,
+      "eval_precision": 0.7933178666588052,
+      "eval_recall": 0.5918231315905315,
+      "eval_runtime": 8.8153,
+      "eval_samples_per_second": 77.025,
+      "eval_steps_per_second": 2.496,
+      "step": 47
+    },
+    {
+      "epoch": 1.0471204188481675,
+      "grad_norm": 1.8398358821868896,
+      "learning_rate": 4.383886255924171e-05,
+      "loss": 0.5637,
+      "step": 50
+    },
+    {
+      "epoch": 1.256544502617801,
+      "grad_norm": 1.4762039184570312,
+      "learning_rate": 4.146919431279621e-05,
+      "loss": 0.4938,
+      "step": 60
+    },
+    {
+      "epoch": 1.4659685863874345,
+      "grad_norm": 2.8683714866638184,
+      "learning_rate": 3.909952606635071e-05,
+      "loss": 0.477,
+      "step": 70
+    },
+    {
+      "epoch": 1.675392670157068,
+      "grad_norm": 1.5742051601409912,
+      "learning_rate": 3.672985781990522e-05,
+      "loss": 0.4335,
+      "step": 80
+    },
+    {
+      "epoch": 1.8848167539267016,
+      "grad_norm": 2.7039575576782227,
+      "learning_rate": 3.4360189573459716e-05,
+      "loss": 0.4486,
+      "step": 90
+    },
+    {
+      "epoch": 1.9895287958115184,
+      "eval_accuracy": 0.801178203240059,
+      "eval_f1": 0.7213482384744982,
+      "eval_loss": 0.46302151679992676,
+      "eval_precision": 0.7964021663408166,
+      "eval_recall": 0.682381066062999,
+      "eval_runtime": 8.91,
+      "eval_samples_per_second": 76.206,
+      "eval_steps_per_second": 2.469,
+      "step": 95
+    },
+    {
+      "epoch": 2.094240837696335,
+      "grad_norm": 2.257824182510376,
+      "learning_rate": 3.1990521327014215e-05,
+      "loss": 0.3847,
+      "step": 100
+    },
+    {
+      "epoch": 2.303664921465969,
+      "grad_norm": 1.9415006637573242,
+      "learning_rate": 2.962085308056872e-05,
+      "loss": 0.3028,
+      "step": 110
+    },
+    {
+      "epoch": 2.513089005235602,
+      "grad_norm": 1.7105799913406372,
+      "learning_rate": 2.7251184834123224e-05,
+      "loss": 0.3271,
+      "step": 120
+    },
+    {
+      "epoch": 2.7225130890052354,
+      "grad_norm": 2.8020598888397217,
+      "learning_rate": 2.4881516587677726e-05,
+      "loss": 0.3494,
+      "step": 130
+    },
+    {
+      "epoch": 2.931937172774869,
+      "grad_norm": 1.7589322328567505,
+      "learning_rate": 2.251184834123223e-05,
+      "loss": 0.3285,
+      "step": 140
+    },
+    {
+      "epoch": 2.994764397905759,
+      "eval_accuracy": 0.8291605301914581,
+      "eval_f1": 0.7721255592297218,
+      "eval_loss": 0.4393855333328247,
+      "eval_precision": 0.823206019483298,
+      "eval_recall": 0.7366059517474952,
+      "eval_runtime": 9.0269,
+      "eval_samples_per_second": 75.219,
+      "eval_steps_per_second": 2.437,
+      "step": 143
+    },
+    {
+      "epoch": 3.141361256544503,
+      "grad_norm": 1.3418846130371094,
+      "learning_rate": 2.014218009478673e-05,
+      "loss": 0.2621,
+      "step": 150
+    },
+    {
+      "epoch": 3.350785340314136,
+      "grad_norm": 1.225538969039917,
+      "learning_rate": 1.7772511848341233e-05,
+      "loss": 0.2541,
+      "step": 160
+    },
+    {
+      "epoch": 3.5602094240837694,
+      "grad_norm": 1.6284271478652954,
+      "learning_rate": 1.5402843601895736e-05,
+      "loss": 0.2234,
+      "step": 170
+    },
+    {
+      "epoch": 3.769633507853403,
+      "grad_norm": 1.5704679489135742,
+      "learning_rate": 1.3033175355450238e-05,
+      "loss": 0.2444,
+      "step": 180
+    },
+    {
+      "epoch": 3.979057591623037,
+      "grad_norm": 1.8943076133728027,
+      "learning_rate": 1.066350710900474e-05,
+      "loss": 0.2391,
+      "step": 190
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8114874815905744,
+      "eval_f1": 0.7554988545936505,
+      "eval_loss": 0.43024498224258423,
+      "eval_precision": 0.7941367088101344,
+      "eval_recall": 0.7332808654662242,
+      "eval_runtime": 8.9879,
+      "eval_samples_per_second": 75.546,
+      "eval_steps_per_second": 2.448,
+      "step": 191
+    },
+    {
+      "epoch": 4.18848167539267,
+      "grad_norm": 2.159135580062866,
+      "learning_rate": 8.293838862559241e-06,
+      "loss": 0.187,
+      "step": 200
+    },
+    {
+      "epoch": 4.397905759162303,
+      "grad_norm": 1.2371037006378174,
+      "learning_rate": 5.924170616113745e-06,
+      "loss": 0.1948,
+      "step": 210
+    },
+    {
+      "epoch": 4.607329842931938,
+      "grad_norm": 1.2540051937103271,
+      "learning_rate": 3.5545023696682464e-06,
+      "loss": 0.1777,
+      "step": 220
+    },
+    {
+      "epoch": 4.816753926701571,
+      "grad_norm": 1.5002622604370117,
+      "learning_rate": 1.1848341232227488e-06,
+      "loss": 0.1814,
+      "step": 230
+    },
+    {
+      "epoch": 4.9214659685863875,
+      "eval_accuracy": 0.8217967599410898,
+      "eval_f1": 0.7631453012448455,
+      "eval_loss": 0.43649259209632874,
+      "eval_precision": 0.7993023457173745,
+      "eval_recall": 0.7362480354321321,
+      "eval_runtime": 9.0379,
+      "eval_samples_per_second": 75.128,
+      "eval_steps_per_second": 2.434,
+      "step": 235
+    },
+    {
+      "epoch": 4.9214659685863875,
+      "step": 235,
+      "total_flos": 2.328211069271507e+18,
+      "train_loss": 0.4179329159411978,
+      "train_runtime": 724.011,
+      "train_samples_per_second": 42.147,
+      "train_steps_per_second": 0.325
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 235,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 2.328211069271507e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88af6720641198e2070d53c4960a14afe3614305500225d61695e28818257076
+size 5048