End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.9904761904761905,
-    "eval_loss": 0.08451159298419952,
-    "eval_runtime": 1.8446,
-    "eval_samples_per_second": 113.843,
-    "eval_steps_per_second": 3.795,
-    "total_flos": 4.697913185667072e+16,
-    "train_loss": 0.6861413399378459,
-    "train_runtime": 27.5947,
-    "train_samples_per_second": 342.458,
-    "train_steps_per_second": 2.718
+    "epoch": 2.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.008575703017413616,
+    "eval_runtime": 1.7087,
+    "eval_samples_per_second": 122.9,
+    "eval_steps_per_second": 4.097,
+    "total_flos": 9.395826371334144e+16,
+    "train_loss": 0.33677051266034447,
+    "train_runtime": 52.9479,
+    "train_samples_per_second": 178.477,
+    "train_steps_per_second": 1.416
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_accuracy": 0.9904761904761905,
-    "eval_loss": 0.08451159298419952,
-    "eval_runtime": 1.8446,
-    "eval_samples_per_second": 113.843,
-    "eval_steps_per_second": 3.795
+    "epoch": 2.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.008575703017413616,
+    "eval_runtime": 1.7087,
+    "eval_samples_per_second": 122.9,
+    "eval_steps_per_second": 4.097
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "total_flos": 4.697913185667072e+16,
-    "train_loss": 0.6861413399378459,
-    "train_runtime": 27.5947,
-    "train_samples_per_second": 342.458,
-    "train_steps_per_second": 2.718
+    "epoch": 2.0,
+    "total_flos": 9.395826371334144e+16,
+    "train_loss": 0.33677051266034447,
+    "train_runtime": 52.9479,
+    "train_samples_per_second": 178.477,
+    "train_steps_per_second": 1.416
 }

trainer_state.json

@@ -1,37 +1,60 @@
 {
-  "best_metric": 0.9904761904761905,
-  "best_model_checkpoint": "/kaggle/working/swin-brain-modality-classification/checkpoint-15",
-  "epoch": 1.0,
+  "best_metric": 1.0,
+  "best_model_checkpoint": "/kaggle/working/swin-brain-modality-classification/checkpoint-30",
+  "epoch": 2.0,
   "eval_steps": 500,
-  "global_step": 15,
+  "global_step": 30,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.6666666666666666,
-      "grad_norm": 12.149126052856445,
+      "grad_norm": 5.258427619934082,
       "learning_rate": 4.850746268656717e-05,
-      "loss": 0.9131,
+      "loss": 0.8454,
       "step": 10
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9904761904761905,
-      "eval_loss": 0.08451159298419952,
-      "eval_runtime": 1.8249,
-      "eval_samples_per_second": 115.074,
-      "eval_steps_per_second": 3.836,
+      "eval_accuracy": 0.9761904761904762,
+      "eval_loss": 0.11262203752994537,
+      "eval_runtime": 1.7281,
+      "eval_samples_per_second": 121.524,
+      "eval_steps_per_second": 4.051,
       "step": 15
     },
     {
-      "epoch": 1.0,
-      "step": 15,
-      "total_flos": 4.697913185667072e+16,
-      "train_loss": 0.6861413399378459,
-      "train_runtime": 27.5947,
-      "train_samples_per_second": 342.458,
-      "train_steps_per_second": 2.718
+      "epoch": 1.3333333333333333,
+      "grad_norm": 3.9709157943725586,
+      "learning_rate": 4.104477611940299e-05,
+      "loss": 0.1186,
+      "step": 20
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.1410547494888306,
+      "learning_rate": 3.358208955223881e-05,
+      "loss": 0.0463,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.008575703017413616,
+      "eval_runtime": 1.7615,
+      "eval_samples_per_second": 119.214,
+      "eval_steps_per_second": 3.974,
+      "step": 30
+    },
+    {
+      "epoch": 2.0,
+      "step": 30,
+      "total_flos": 9.395826371334144e+16,
+      "train_loss": 0.33677051266034447,
+      "train_runtime": 52.9479,
+      "train_samples_per_second": 178.477,
+      "train_steps_per_second": 1.416
     }
   ],
   "logging_steps": 10,
@@ -60,7 +83,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.697913185667072e+16,
+  "total_flos": 9.395826371334144e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null