End of training

README.md

@@ -30,7 +30,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5949
+- Loss: 0.5859
 - Accuracy: 0.7442
 
 ## Model description

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 9.6,
-    "eval_accuracy": 0.813953488372093,
-    "eval_loss": 0.4812479317188263,
-    "eval_runtime": 0.2934,
-    "eval_samples_per_second": 146.538,
-    "eval_steps_per_second": 37.487,
+    "eval_accuracy": 0.7441860465116279,
+    "eval_loss": 0.5859283208847046,
+    "eval_runtime": 0.3168,
+    "eval_samples_per_second": 135.736,
+    "eval_steps_per_second": 34.723,
     "total_flos": 7.3694882123477e+16,
-    "train_loss": 0.5894826014836629,
-    "train_runtime": 33.2904,
-    "train_samples_per_second": 29.738,
-    "train_steps_per_second": 1.802
+    "train_loss": 0.5492916504542033,
+    "train_runtime": 34.9653,
+    "train_samples_per_second": 28.314,
+    "train_steps_per_second": 1.716
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.6,
-    "eval_accuracy": 0.813953488372093,
-    "eval_loss": 0.4812479317188263,
-    "eval_runtime": 0.2934,
-    "eval_samples_per_second": 146.538,
-    "eval_steps_per_second": 37.487
+    "eval_accuracy": 0.7441860465116279,
+    "eval_loss": 0.5859283208847046,
+    "eval_runtime": 0.3168,
+    "eval_samples_per_second": 135.736,
+    "eval_steps_per_second": 34.723
 }

runs/Aug10_00-06-07_EMIMDGXA100GPU1/events.out.tfevents.1723262811.EMIMDGXA100GPU1.4103107.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d323b6c860fa046542c65ec72fed6cdb31109a806e6af201fb12ef6175f023ec
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 9.6,
     "total_flos": 7.3694882123477e+16,
-    "train_loss": 0.5894826014836629,
-    "train_runtime": 33.2904,
-    "train_samples_per_second": 29.738,
-    "train_steps_per_second": 1.802
+    "train_loss": 0.5492916504542033,
+    "train_runtime": 34.9653,
+    "train_samples_per_second": 28.314,
+    "train_steps_per_second": 1.716
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.813953488372093,
-  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-12",
+  "best_metric": 0.7441860465116279,
+  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-6",
   "epoch": 9.6,
   "eval_steps": 500,
   "global_step": 60,
@@ -10,144 +10,144 @@
   "log_history": [
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.7906976744186046,
-      "eval_loss": 0.4999209940433502,
-      "eval_runtime": 0.2683,
-      "eval_samples_per_second": 160.24,
-      "eval_steps_per_second": 40.992,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5859283208847046,
+      "eval_runtime": 0.2566,
+      "eval_samples_per_second": 167.599,
+      "eval_steps_per_second": 42.874,
       "step": 6
     },
     {
       "epoch": 1.6,
-      "grad_norm": 3.087608575820923,
+      "grad_norm": 3.7175180912017822,
       "learning_rate": 4.62962962962963e-06,
-      "loss": 0.5993,
+      "loss": 0.605,
       "step": 10
     },
     {
       "epoch": 1.92,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.4812479317188263,
-      "eval_runtime": 0.3038,
-      "eval_samples_per_second": 141.519,
-      "eval_steps_per_second": 36.202,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5841977000236511,
+      "eval_runtime": 0.2709,
+      "eval_samples_per_second": 158.721,
+      "eval_steps_per_second": 40.603,
       "step": 12
     },
     {
       "epoch": 2.88,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.48915746808052063,
-      "eval_runtime": 0.2696,
-      "eval_samples_per_second": 159.518,
-      "eval_steps_per_second": 40.807,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5918598771095276,
+      "eval_runtime": 0.2627,
+      "eval_samples_per_second": 163.663,
+      "eval_steps_per_second": 41.867,
       "step": 18
     },
     {
       "epoch": 3.2,
-      "grad_norm": 5.1914448738098145,
+      "grad_norm": 2.415996551513672,
       "learning_rate": 3.7037037037037037e-06,
-      "loss": 0.6031,
+      "loss": 0.5428,
       "step": 20
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.49750375747680664,
-      "eval_runtime": 0.2506,
-      "eval_samples_per_second": 171.576,
-      "eval_steps_per_second": 43.892,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5884882807731628,
+      "eval_runtime": 0.9618,
+      "eval_samples_per_second": 44.706,
+      "eval_steps_per_second": 11.436,
       "step": 25
     },
     {
       "epoch": 4.8,
-      "grad_norm": 4.207640171051025,
+      "grad_norm": 2.7047529220581055,
       "learning_rate": 2.7777777777777783e-06,
-      "loss": 0.5946,
+      "loss": 0.5584,
       "step": 30
     },
     {
       "epoch": 4.96,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5062411427497864,
-      "eval_runtime": 0.263,
-      "eval_samples_per_second": 163.481,
-      "eval_steps_per_second": 41.821,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.588610827922821,
+      "eval_runtime": 0.2732,
+      "eval_samples_per_second": 157.373,
+      "eval_steps_per_second": 40.258,
       "step": 31
     },
     {
       "epoch": 5.92,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5119370222091675,
-      "eval_runtime": 0.2714,
-      "eval_samples_per_second": 158.421,
-      "eval_steps_per_second": 40.526,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5914657711982727,
+      "eval_runtime": 0.2503,
+      "eval_samples_per_second": 171.764,
+      "eval_steps_per_second": 43.94,
       "step": 37
     },
     {
       "epoch": 6.4,
-      "grad_norm": 4.061374664306641,
+      "grad_norm": 3.9282095432281494,
       "learning_rate": 1.8518518518518519e-06,
-      "loss": 0.5557,
+      "loss": 0.5593,
       "step": 40
     },
     {
       "epoch": 6.88,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5102289915084839,
-      "eval_runtime": 0.2732,
-      "eval_samples_per_second": 157.394,
-      "eval_steps_per_second": 40.264,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5934743881225586,
+      "eval_runtime": 0.2654,
+      "eval_samples_per_second": 162.014,
+      "eval_steps_per_second": 41.446,
       "step": 43
     },
     {
       "epoch": 8.0,
-      "grad_norm": 3.359290361404419,
+      "grad_norm": 2.82995343208313,
       "learning_rate": 9.259259259259259e-07,
-      "loss": 0.583,
+      "loss": 0.5097,
       "step": 50
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5075170993804932,
-      "eval_runtime": 0.2683,
-      "eval_samples_per_second": 160.255,
-      "eval_steps_per_second": 40.996,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5947032570838928,
+      "eval_runtime": 0.2582,
+      "eval_samples_per_second": 166.523,
+      "eval_steps_per_second": 42.599,
       "step": 50
     },
     {
       "epoch": 8.96,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5080327987670898,
-      "eval_runtime": 0.2647,
-      "eval_samples_per_second": 162.473,
-      "eval_steps_per_second": 41.563,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.594927966594696,
+      "eval_runtime": 0.2847,
+      "eval_samples_per_second": 151.012,
+      "eval_steps_per_second": 38.631,
       "step": 56
     },
     {
       "epoch": 9.6,
-      "grad_norm": 4.910851001739502,
+      "grad_norm": 3.3591854572296143,
       "learning_rate": 0.0,
-      "loss": 0.6013,
+      "loss": 0.5205,
       "step": 60
     },
     {
       "epoch": 9.6,
-      "eval_accuracy": 0.813953488372093,
-      "eval_loss": 0.5078577399253845,
-      "eval_runtime": 0.2596,
-      "eval_samples_per_second": 165.63,
-      "eval_steps_per_second": 42.371,
+      "eval_accuracy": 0.7441860465116279,
+      "eval_loss": 0.5949187874794006,
+      "eval_runtime": 0.2565,
+      "eval_samples_per_second": 167.627,
+      "eval_steps_per_second": 42.881,
       "step": 60
     },
     {
       "epoch": 9.6,
       "step": 60,
       "total_flos": 7.3694882123477e+16,
-      "train_loss": 0.5894826014836629,
-      "train_runtime": 33.2904,
-      "train_samples_per_second": 29.738,
-      "train_steps_per_second": 1.802
+      "train_loss": 0.5492916504542033,
+      "train_runtime": 34.9653,
+      "train_samples_per_second": 28.314,
+      "train_steps_per_second": 1.716
     }
   ],
   "logging_steps": 10,