End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3080
-- Accuracy: 0.8619
+- Loss: 0.3120
+- Accuracy: 0.8661
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 1.9555555555555557,
-    "eval_accuracy": 0.897489539748954,
-    "eval_loss": 0.23102231323719025,
-    "eval_runtime": 2.1759,
-    "eval_samples_per_second": 219.675,
-    "eval_steps_per_second": 6.894,
+    "eval_accuracy": 0.8661087866108786,
+    "eval_loss": 0.31204754114151,
+    "eval_runtime": 2.1777,
+    "eval_samples_per_second": 219.5,
+    "eval_steps_per_second": 6.888,
     "total_flos": 2.0943656319482266e+17,
-    "train_loss": 0.3058708075321082,
-    "train_runtime": 94.9406,
-    "train_samples_per_second": 90.541,
-    "train_steps_per_second": 0.695
+    "train_loss": 0.31210848598769214,
+    "train_runtime": 94.7159,
+    "train_samples_per_second": 90.756,
+    "train_steps_per_second": 0.697
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.9555555555555557,
-    "eval_accuracy": 0.897489539748954,
-    "eval_loss": 0.23102231323719025,
-    "eval_runtime": 2.1759,
-    "eval_samples_per_second": 219.675,
-    "eval_steps_per_second": 6.894
+    "eval_accuracy": 0.8661087866108786,
+    "eval_loss": 0.31204754114151,
+    "eval_runtime": 2.1777,
+    "eval_samples_per_second": 219.5,
+    "eval_steps_per_second": 6.888
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.9555555555555557,
     "total_flos": 2.0943656319482266e+17,
-    "train_loss": 0.3058708075321082,
-    "train_runtime": 94.9406,
-    "train_samples_per_second": 90.541,
-    "train_steps_per_second": 0.695
+    "train_loss": 0.31210848598769214,
+    "train_runtime": 94.7159,
+    "train_samples_per_second": 90.756,
+    "train_steps_per_second": 0.697
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.897489539748954,
+  "best_metric": 0.8661087866108786,
   "best_model_checkpoint": "/kaggle/working/swin-brain-abnormalities-classification/checkpoint-33",
   "epoch": 1.9555555555555557,
   "eval_steps": 500,
@@ -10,72 +10,72 @@
   "log_history": [
     {
       "epoch": 0.2962962962962963,
-      "grad_norm": 3.896357297897339,
+      "grad_norm": 4.365435600280762,
       "learning_rate": 4.745762711864407e-05,
-      "loss": 0.6251,
+      "loss": 0.6291,
       "step": 10
     },
     {
       "epoch": 0.5925925925925926,
-      "grad_norm": 8.837748527526855,
+      "grad_norm": 5.855260848999023,
       "learning_rate": 3.898305084745763e-05,
-      "loss": 0.38,
+      "loss": 0.4017,
       "step": 20
     },
     {
       "epoch": 0.8888888888888888,
-      "grad_norm": 9.348395347595215,
+      "grad_norm": 5.322465896606445,
       "learning_rate": 3.050847457627119e-05,
-      "loss": 0.2769,
+      "loss": 0.2865,
       "step": 30
     },
     {
       "epoch": 0.9777777777777777,
-      "eval_accuracy": 0.897489539748954,
-      "eval_loss": 0.23102231323719025,
-      "eval_runtime": 2.1587,
-      "eval_samples_per_second": 221.433,
-      "eval_steps_per_second": 6.949,
+      "eval_accuracy": 0.8661087866108786,
+      "eval_loss": 0.31204754114151,
+      "eval_runtime": 2.1954,
+      "eval_samples_per_second": 217.726,
+      "eval_steps_per_second": 6.832,
       "step": 33
     },
     {
       "epoch": 1.1851851851851851,
-      "grad_norm": 8.096166610717773,
+      "grad_norm": 5.393370151519775,
       "learning_rate": 2.2033898305084748e-05,
-      "loss": 0.2364,
+      "loss": 0.2308,
       "step": 40
     },
     {
       "epoch": 1.4814814814814814,
-      "grad_norm": 15.118054389953613,
+      "grad_norm": 14.526658058166504,
       "learning_rate": 1.3559322033898305e-05,
-      "loss": 0.2129,
+      "loss": 0.2127,
       "step": 50
     },
     {
       "epoch": 1.7777777777777777,
-      "grad_norm": 6.056334495544434,
+      "grad_norm": 6.8975911140441895,
       "learning_rate": 5.084745762711865e-06,
-      "loss": 0.1861,
+      "loss": 0.1943,
       "step": 60
     },
     {
       "epoch": 1.9555555555555557,
-      "eval_accuracy": 0.897489539748954,
-      "eval_loss": 0.24522443115711212,
-      "eval_runtime": 2.2548,
-      "eval_samples_per_second": 211.994,
-      "eval_steps_per_second": 6.653,
+      "eval_accuracy": 0.8619246861924686,
+      "eval_loss": 0.3080492317676544,
+      "eval_runtime": 2.2579,
+      "eval_samples_per_second": 211.701,
+      "eval_steps_per_second": 6.643,
       "step": 66
     },
     {
       "epoch": 1.9555555555555557,
       "step": 66,
       "total_flos": 2.0943656319482266e+17,
-      "train_loss": 0.3058708075321082,
-      "train_runtime": 94.9406,
-      "train_samples_per_second": 90.541,
-      "train_steps_per_second": 0.695
+      "train_loss": 0.31210848598769214,
+      "train_runtime": 94.7159,
+      "train_samples_per_second": 90.756,
+      "train_steps_per_second": 0.697
     }
   ],
   "logging_steps": 10,