End of training

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8705882352941177
+      value: 0.9235294117647059
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/vit-msn-small](https://huggingface.co/facebook/vit-msn-small) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3307
-- Accuracy: 0.8706
+- Loss: 0.1866
+- Accuracy: 0.9235
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.9235294117647059,
+    "eval_loss": 0.18661858141422272,
+    "eval_runtime": 0.574,
+    "eval_samples_per_second": 296.153,
+    "eval_steps_per_second": 10.452,
+    "total_flos": 2.9117823636013056e+17,
+    "train_loss": 0.28690393567085265,
+    "train_runtime": 133.8542,
+    "train_samples_per_second": 111.166,
+    "train_steps_per_second": 0.896
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.9235294117647059,
+    "eval_loss": 0.18661858141422272,
+    "eval_runtime": 0.574,
+    "eval_samples_per_second": 296.153,
+    "eval_steps_per_second": 10.452
+}

runs/Feb21_15-54-52_3abf47ac437f/events.out.tfevents.1740153459.3abf47ac437f.209.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d545e38339fe97c9c4a0b7b2f31e0c654ccdcbdbe59d70f68edfa5c5fc776b2
+size 405

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 20.0,
+    "total_flos": 2.9117823636013056e+17,
+    "train_loss": 0.28690393567085265,
+    "train_runtime": 133.8542,
+    "train_samples_per_second": 111.166,
+    "train_steps_per_second": 0.896
+}

trainer_state.json

@@ -0,0 +1,306 @@
+{
+  "best_metric": 0.9235294117647059,
+  "best_model_checkpoint": "vit-msn-small-lipid-invalidation-nobg/checkpoint-84",
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7058823529411765,
+      "eval_loss": 0.5867760181427002,
+      "eval_runtime": 0.536,
+      "eval_samples_per_second": 317.18,
+      "eval_steps_per_second": 11.195,
+      "step": 6
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 7.053527355194092,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.583,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.8705882352941177,
+      "eval_loss": 0.28499943017959595,
+      "eval_runtime": 0.5262,
+      "eval_samples_per_second": 323.046,
+      "eval_steps_per_second": 11.402,
+      "step": 12
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.8647058823529412,
+      "eval_loss": 0.31667008996009827,
+      "eval_runtime": 0.5367,
+      "eval_samples_per_second": 316.749,
+      "eval_steps_per_second": 11.179,
+      "step": 18
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 8.924991607666016,
+      "learning_rate": 4.62962962962963e-05,
+      "loss": 0.4996,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8294117647058824,
+      "eval_loss": 0.37376630306243896,
+      "eval_runtime": 0.5648,
+      "eval_samples_per_second": 301.01,
+      "eval_steps_per_second": 10.624,
+      "step": 24
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 9.494657516479492,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.3097,
+      "step": 30
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9176470588235294,
+      "eval_loss": 0.21050500869750977,
+      "eval_runtime": 0.5568,
+      "eval_samples_per_second": 305.341,
+      "eval_steps_per_second": 10.777,
+      "step": 30
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8823529411764706,
+      "eval_loss": 0.26954787969589233,
+      "eval_runtime": 0.539,
+      "eval_samples_per_second": 315.42,
+      "eval_steps_per_second": 11.132,
+      "step": 36
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 10.003211975097656,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.2893,
+      "step": 40
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.8941176470588236,
+      "eval_loss": 0.23035354912281036,
+      "eval_runtime": 0.5649,
+      "eval_samples_per_second": 300.957,
+      "eval_steps_per_second": 10.622,
+      "step": 42
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.8529411764705882,
+      "eval_loss": 0.297798752784729,
+      "eval_runtime": 0.5553,
+      "eval_samples_per_second": 306.117,
+      "eval_steps_per_second": 10.804,
+      "step": 48
+    },
+    {
+      "epoch": 8.333333333333334,
+      "grad_norm": 5.467005252838135,
+      "learning_rate": 3.240740740740741e-05,
+      "loss": 0.3305,
+      "step": 50
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.888235294117647,
+      "eval_loss": 0.2952236533164978,
+      "eval_runtime": 0.5508,
+      "eval_samples_per_second": 308.654,
+      "eval_steps_per_second": 10.894,
+      "step": 54
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 13.88063907623291,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.2775,
+      "step": 60
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.888235294117647,
+      "eval_loss": 0.2661350965499878,
+      "eval_runtime": 0.5605,
+      "eval_samples_per_second": 303.291,
+      "eval_steps_per_second": 10.704,
+      "step": 60
+    },
+    {
+      "epoch": 11.0,
+      "eval_accuracy": 0.8823529411764706,
+      "eval_loss": 0.25326403975486755,
+      "eval_runtime": 0.5723,
+      "eval_samples_per_second": 297.046,
+      "eval_steps_per_second": 10.484,
+      "step": 66
+    },
+    {
+      "epoch": 11.666666666666666,
+      "grad_norm": 7.95721960067749,
+      "learning_rate": 2.314814814814815e-05,
+      "loss": 0.223,
+      "step": 70
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.888235294117647,
+      "eval_loss": 0.2585911154747009,
+      "eval_runtime": 0.5681,
+      "eval_samples_per_second": 299.222,
+      "eval_steps_per_second": 10.561,
+      "step": 72
+    },
+    {
+      "epoch": 13.0,
+      "eval_accuracy": 0.9176470588235294,
+      "eval_loss": 0.2033568024635315,
+      "eval_runtime": 0.5439,
+      "eval_samples_per_second": 312.563,
+      "eval_steps_per_second": 11.032,
+      "step": 78
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 10.242790222167969,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.216,
+      "step": 80
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.9235294117647059,
+      "eval_loss": 0.18661858141422272,
+      "eval_runtime": 0.5426,
+      "eval_samples_per_second": 313.32,
+      "eval_steps_per_second": 11.058,
+      "step": 84
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 5.154911518096924,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.1858,
+      "step": 90
+    },
+    {
+      "epoch": 15.0,
+      "eval_accuracy": 0.888235294117647,
+      "eval_loss": 0.2631073296070099,
+      "eval_runtime": 0.5262,
+      "eval_samples_per_second": 323.067,
+      "eval_steps_per_second": 11.402,
+      "step": 90
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.888235294117647,
+      "eval_loss": 0.2439642697572708,
+      "eval_runtime": 0.5331,
+      "eval_samples_per_second": 318.919,
+      "eval_steps_per_second": 11.256,
+      "step": 96
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 3.7305550575256348,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.1866,
+      "step": 100
+    },
+    {
+      "epoch": 17.0,
+      "eval_accuracy": 0.8941176470588236,
+      "eval_loss": 0.2479913830757141,
+      "eval_runtime": 0.6023,
+      "eval_samples_per_second": 282.236,
+      "eval_steps_per_second": 9.961,
+      "step": 102
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.8647058823529412,
+      "eval_loss": 0.38971054553985596,
+      "eval_runtime": 0.5781,
+      "eval_samples_per_second": 294.055,
+      "eval_steps_per_second": 10.378,
+      "step": 108
+    },
+    {
+      "epoch": 18.333333333333332,
+      "grad_norm": 16.46074676513672,
+      "learning_rate": 4.6296296296296296e-06,
+      "loss": 0.1701,
+      "step": 110
+    },
+    {
+      "epoch": 19.0,
+      "eval_accuracy": 0.8647058823529412,
+      "eval_loss": 0.3998810350894928,
+      "eval_runtime": 0.5376,
+      "eval_samples_per_second": 316.242,
+      "eval_steps_per_second": 11.161,
+      "step": 114
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 5.165961265563965,
+      "learning_rate": 0.0,
+      "loss": 0.1717,
+      "step": 120
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.8705882352941177,
+      "eval_loss": 0.33069366216659546,
+      "eval_runtime": 0.5592,
+      "eval_samples_per_second": 304.018,
+      "eval_steps_per_second": 10.73,
+      "step": 120
+    },
+    {
+      "epoch": 20.0,
+      "step": 120,
+      "total_flos": 2.9117823636013056e+17,
+      "train_loss": 0.28690393567085265,
+      "train_runtime": 133.8542,
+      "train_samples_per_second": 111.166,
+      "train_steps_per_second": 0.896
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 120,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9117823636013056e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}