End of training

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 4.888888888888889,
+    "eval_accuracy": 0.9665271966527197,
+    "eval_loss": 0.08974912762641907,
+    "eval_runtime": 2.1958,
+    "eval_samples_per_second": 217.687,
+    "eval_steps_per_second": 6.831,
+    "total_flos": 5.227711603505234e+17,
+    "train_loss": 0.1524324781966932,
+    "train_runtime": 238.6908,
+    "train_samples_per_second": 90.033,
+    "train_steps_per_second": 0.691
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.888888888888889,
+    "eval_accuracy": 0.9665271966527197,
+    "eval_loss": 0.08974912762641907,
+    "eval_runtime": 2.1958,
+    "eval_samples_per_second": 217.687,
+    "eval_steps_per_second": 6.831
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.888888888888889,
+    "total_flos": 5.227711603505234e+17,
+    "train_loss": 0.1524324781966932,
+    "train_runtime": 238.6908,
+    "train_samples_per_second": 90.033,
+    "train_steps_per_second": 0.691
+}

trainer_state.json

@@ -0,0 +1,208 @@
+{
+  "best_metric": 0.9665271966527197,
+  "best_model_checkpoint": "/kaggle/working/swin-brain-abnormalities-classification/checkpoint-165",
+  "epoch": 4.888888888888889,
+  "eval_steps": 500,
+  "global_step": 165,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2962962962962963,
+      "grad_norm": 5.920434951782227,
+      "learning_rate": 2.9411764705882354e-05,
+      "loss": 0.2993,
+      "step": 10
+    },
+    {
+      "epoch": 0.5925925925925926,
+      "grad_norm": 11.511192321777344,
+      "learning_rate": 4.8986486486486486e-05,
+      "loss": 0.2762,
+      "step": 20
+    },
+    {
+      "epoch": 0.8888888888888888,
+      "grad_norm": 11.25218677520752,
+      "learning_rate": 4.560810810810811e-05,
+      "loss": 0.2275,
+      "step": 30
+    },
+    {
+      "epoch": 0.9777777777777777,
+      "eval_accuracy": 0.8786610878661087,
+      "eval_loss": 0.3053434491157532,
+      "eval_runtime": 2.1724,
+      "eval_samples_per_second": 220.033,
+      "eval_steps_per_second": 6.905,
+      "step": 33
+    },
+    {
+      "epoch": 1.1851851851851851,
+      "grad_norm": 7.5714945793151855,
+      "learning_rate": 4.222972972972973e-05,
+      "loss": 0.2104,
+      "step": 40
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 5.060845851898193,
+      "learning_rate": 3.885135135135135e-05,
+      "loss": 0.1783,
+      "step": 50
+    },
+    {
+      "epoch": 1.7777777777777777,
+      "grad_norm": 4.324786186218262,
+      "learning_rate": 3.547297297297297e-05,
+      "loss": 0.1635,
+      "step": 60
+    },
+    {
+      "epoch": 1.9851851851851852,
+      "eval_accuracy": 0.9560669456066946,
+      "eval_loss": 0.1382259577512741,
+      "eval_runtime": 2.1943,
+      "eval_samples_per_second": 217.84,
+      "eval_steps_per_second": 6.836,
+      "step": 67
+    },
+    {
+      "epoch": 2.074074074074074,
+      "grad_norm": 4.530977249145508,
+      "learning_rate": 3.20945945945946e-05,
+      "loss": 0.1554,
+      "step": 70
+    },
+    {
+      "epoch": 2.3703703703703702,
+      "grad_norm": 6.494528770446777,
+      "learning_rate": 2.8716216216216217e-05,
+      "loss": 0.1448,
+      "step": 80
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 4.873690128326416,
+      "learning_rate": 2.533783783783784e-05,
+      "loss": 0.098,
+      "step": 90
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 2.7952539920806885,
+      "learning_rate": 2.195945945945946e-05,
+      "loss": 0.1359,
+      "step": 100
+    },
+    {
+      "epoch": 2.9925925925925925,
+      "eval_accuracy": 0.9205020920502092,
+      "eval_loss": 0.20204488933086395,
+      "eval_runtime": 2.1866,
+      "eval_samples_per_second": 218.608,
+      "eval_steps_per_second": 6.86,
+      "step": 101
+    },
+    {
+      "epoch": 3.259259259259259,
+      "grad_norm": 6.09904670715332,
+      "learning_rate": 1.8581081081081082e-05,
+      "loss": 0.1232,
+      "step": 110
+    },
+    {
+      "epoch": 3.5555555555555554,
+      "grad_norm": 8.267931938171387,
+      "learning_rate": 1.5202702702702704e-05,
+      "loss": 0.1143,
+      "step": 120
+    },
+    {
+      "epoch": 3.851851851851852,
+      "grad_norm": 3.9983508586883545,
+      "learning_rate": 1.1824324324324325e-05,
+      "loss": 0.0876,
+      "step": 130
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9602510460251046,
+      "eval_loss": 0.10952634364366531,
+      "eval_runtime": 2.1565,
+      "eval_samples_per_second": 221.657,
+      "eval_steps_per_second": 6.956,
+      "step": 135
+    },
+    {
+      "epoch": 4.148148148148148,
+      "grad_norm": 8.512967109680176,
+      "learning_rate": 8.445945945945946e-06,
+      "loss": 0.0823,
+      "step": 140
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 5.426952362060547,
+      "learning_rate": 5.067567567567568e-06,
+      "loss": 0.0878,
+      "step": 150
+    },
+    {
+      "epoch": 4.7407407407407405,
+      "grad_norm": 3.7845118045806885,
+      "learning_rate": 1.6891891891891894e-06,
+      "loss": 0.0792,
+      "step": 160
+    },
+    {
+      "epoch": 4.888888888888889,
+      "eval_accuracy": 0.9665271966527197,
+      "eval_loss": 0.08974912762641907,
+      "eval_runtime": 2.3398,
+      "eval_samples_per_second": 204.294,
+      "eval_steps_per_second": 6.411,
+      "step": 165
+    },
+    {
+      "epoch": 4.888888888888889,
+      "step": 165,
+      "total_flos": 5.227711603505234e+17,
+      "train_loss": 0.1524324781966932,
+      "train_runtime": 238.6908,
+      "train_samples_per_second": 90.033,
+      "train_steps_per_second": 0.691
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 165,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "CustomEarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 1,
+        "early_stopping_threshold": 0.0
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 0
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.227711603505234e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}