End of training

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.888888888888889,
-    "eval_accuracy": 0.9841269841269841,
-    "eval_loss": 0.20399044454097748,
-    "eval_runtime": 0.3813,
-    "eval_samples_per_second": 165.246,
-    "eval_steps_per_second": 5.246,
-    "total_flos": 4.228121867100365e+16,
-    "train_loss": 0.7618867800785944,
-    "train_runtime": 21.7408,
-    "train_samples_per_second": 130.4,
-    "train_steps_per_second": 0.92
+    "epoch": 4.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.19110850989818573,
+    "eval_runtime": 0.3814,
+    "eval_samples_per_second": 165.197,
+    "eval_steps_per_second": 5.244,
+    "total_flos": 5.637495822800486e+16,
+    "train_loss": 0.6830070416132609,
+    "train_runtime": 28.7573,
+    "train_samples_per_second": 98.584,
+    "train_steps_per_second": 0.695
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.888888888888889,
-    "eval_accuracy": 0.9841269841269841,
-    "eval_loss": 0.20399044454097748,
-    "eval_runtime": 0.3813,
-    "eval_samples_per_second": 165.246,
-    "eval_steps_per_second": 5.246
+    "epoch": 4.0,
+    "eval_accuracy": 1.0,
+    "eval_loss": 0.19110850989818573,
+    "eval_runtime": 0.3814,
+    "eval_samples_per_second": 165.197,
+    "eval_steps_per_second": 5.244
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.888888888888889,
-    "total_flos": 4.228121867100365e+16,
-    "train_loss": 0.7618867800785944,
-    "train_runtime": 21.7408,
-    "train_samples_per_second": 130.4,
-    "train_steps_per_second": 0.92
+    "epoch": 4.0,
+    "total_flos": 5.637495822800486e+16,
+    "train_loss": 0.6830070416132609,
+    "train_runtime": 28.7573,
+    "train_samples_per_second": 98.584,
+    "train_steps_per_second": 0.695
 }

trainer_state.json

@@ -1,55 +1,64 @@
 {
-  "best_metric": 0.9841269841269841,
-  "best_model_checkpoint": "/kaggle/working/swin-brain-plane-classification/checkpoint-13",
-  "epoch": 2.888888888888889,
+  "best_metric": 1.0,
+  "best_model_checkpoint": "/kaggle/working/swin-brain-plane-classification/checkpoint-18",
+  "epoch": 4.0,
   "eval_steps": 500,
-  "global_step": 13,
+  "global_step": 18,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.8888888888888888,
-      "eval_accuracy": 0.8095238095238095,
-      "eval_loss": 0.743109405040741,
-      "eval_runtime": 0.3713,
-      "eval_samples_per_second": 169.663,
-      "eval_steps_per_second": 5.386,
+      "eval_accuracy": 0.9047619047619048,
+      "eval_loss": 0.8267417550086975,
+      "eval_runtime": 0.3659,
+      "eval_samples_per_second": 172.198,
+      "eval_steps_per_second": 5.467,
       "step": 4
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9682539682539683,
-      "eval_loss": 0.3612231910228729,
-      "eval_runtime": 0.3697,
-      "eval_samples_per_second": 170.43,
-      "eval_steps_per_second": 5.41,
+      "eval_loss": 0.45388203859329224,
+      "eval_runtime": 0.3836,
+      "eval_samples_per_second": 164.226,
+      "eval_steps_per_second": 5.214,
       "step": 9
     },
     {
       "epoch": 2.2222222222222223,
-      "grad_norm": 6.062853813171387,
+      "grad_norm": 19.5561466217041,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.8237,
+      "loss": 0.8656,
       "step": 10
     },
     {
       "epoch": 2.888888888888889,
-      "eval_accuracy": 0.9841269841269841,
-      "eval_loss": 0.20399044454097748,
-      "eval_runtime": 0.3636,
-      "eval_samples_per_second": 173.264,
-      "eval_steps_per_second": 5.5,
+      "eval_accuracy": 0.9682539682539683,
+      "eval_loss": 0.30353549122810364,
+      "eval_runtime": 0.368,
+      "eval_samples_per_second": 171.182,
+      "eval_steps_per_second": 5.434,
       "step": 13
     },
     {
-      "epoch": 2.888888888888889,
-      "step": 13,
-      "total_flos": 4.228121867100365e+16,
-      "train_loss": 0.7618867800785944,
-      "train_runtime": 21.7408,
-      "train_samples_per_second": 130.4,
-      "train_steps_per_second": 0.92
+      "epoch": 4.0,
+      "eval_accuracy": 1.0,
+      "eval_loss": 0.19110850989818573,
+      "eval_runtime": 0.3651,
+      "eval_samples_per_second": 172.56,
+      "eval_steps_per_second": 5.478,
+      "step": 18
+    },
+    {
+      "epoch": 4.0,
+      "step": 18,
+      "total_flos": 5.637495822800486e+16,
+      "train_loss": 0.6830070416132609,
+      "train_runtime": 28.7573,
+      "train_samples_per_second": 98.584,
+      "train_steps_per_second": 0.695
     }
   ],
   "logging_steps": 10,
@@ -78,7 +87,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4.228121867100365e+16,
+  "total_flos": 5.637495822800486e+16,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null