End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 2.968421052631579,
-    "eval_accuracy": 0.9544444444444444,
-    "eval_loss": 0.13685034215450287,
-    "eval_runtime": 5.9923,
-    "eval_samples_per_second": 450.581,
-    "eval_steps_per_second": 3.671,
+    "eval_accuracy": 0.9507407407407408,
+    "eval_loss": 0.14795009791851044,
+    "eval_runtime": 5.7904,
+    "eval_samples_per_second": 466.286,
+    "eval_steps_per_second": 3.799,
     "total_flos": 1.7938102147100836e+18,
-    "train_loss": 0.5156102181537777,
-    "train_runtime": 246.7274,
-    "train_samples_per_second": 295.468,
-    "train_steps_per_second": 0.571
+    "train_loss": 0.5108092158821458,
+    "train_runtime": 247.4957,
+    "train_samples_per_second": 294.551,
+    "train_steps_per_second": 0.57
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.968421052631579,
-    "eval_accuracy": 0.9544444444444444,
-    "eval_loss": 0.13685034215450287,
-    "eval_runtime": 5.9923,
-    "eval_samples_per_second": 450.581,
-    "eval_steps_per_second": 3.671
+    "eval_accuracy": 0.9507407407407408,
+    "eval_loss": 0.14795009791851044,
+    "eval_runtime": 5.7904,
+    "eval_samples_per_second": 466.286,
+    "eval_steps_per_second": 3.799
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.968421052631579,
     "total_flos": 1.7938102147100836e+18,
-    "train_loss": 0.5156102181537777,
-    "train_runtime": 246.7274,
-    "train_samples_per_second": 295.468,
-    "train_steps_per_second": 0.571
+    "train_loss": 0.5108092158821458,
+    "train_runtime": 247.4957,
+    "train_samples_per_second": 294.551,
+    "train_steps_per_second": 0.57
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9544444444444444,
+  "best_metric": 0.9507407407407408,
   "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-141",
   "epoch": 2.968421052631579,
   "eval_steps": 500,
@@ -10,137 +10,137 @@
   "log_history": [
     {
       "epoch": 0.21052631578947367,
-      "grad_norm": 3.664080858230591,
+      "grad_norm": 4.6187968254089355,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 2.2003,
+      "loss": 2.2058,
       "step": 10
     },
     {
       "epoch": 0.42105263157894735,
-      "grad_norm": 4.389891147613525,
+      "grad_norm": 13.80441665649414,
       "learning_rate": 4.801587301587302e-05,
-      "loss": 1.3695,
+      "loss": 1.4136,
       "step": 20
     },
     {
       "epoch": 0.631578947368421,
-      "grad_norm": 10.958775520324707,
+      "grad_norm": 8.810165405273438,
       "learning_rate": 4.404761904761905e-05,
-      "loss": 0.6244,
+      "loss": 0.6209,
       "step": 30
     },
     {
       "epoch": 0.8421052631578947,
-      "grad_norm": 9.257582664489746,
+      "grad_norm": 9.584736824035645,
       "learning_rate": 4.007936507936508e-05,
-      "loss": 0.4313,
+      "loss": 0.4418,
       "step": 40
     },
     {
       "epoch": 0.9894736842105263,
-      "eval_accuracy": 0.9166666666666666,
-      "eval_loss": 0.25173959136009216,
-      "eval_runtime": 5.9391,
-      "eval_samples_per_second": 454.618,
-      "eval_steps_per_second": 3.704,
+      "eval_accuracy": 0.9196296296296296,
+      "eval_loss": 0.24092020094394684,
+      "eval_runtime": 6.6105,
+      "eval_samples_per_second": 408.443,
+      "eval_steps_per_second": 3.328,
       "step": 47
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 14.306416511535645,
+      "grad_norm": 6.858447551727295,
       "learning_rate": 3.611111111111111e-05,
-      "loss": 0.3752,
+      "loss": 0.3384,
       "step": 50
     },
     {
       "epoch": 1.263157894736842,
-      "grad_norm": 6.0008978843688965,
+      "grad_norm": 5.806480884552002,
       "learning_rate": 3.2142857142857144e-05,
-      "loss": 0.307,
+      "loss": 0.3164,
       "step": 60
     },
     {
       "epoch": 1.4736842105263157,
-      "grad_norm": 7.767253875732422,
+      "grad_norm": 5.885347366333008,
       "learning_rate": 2.8174603174603175e-05,
-      "loss": 0.291,
+      "loss": 0.267,
       "step": 70
     },
     {
       "epoch": 1.6842105263157894,
-      "grad_norm": 10.404711723327637,
+      "grad_norm": 9.434675216674805,
       "learning_rate": 2.4206349206349206e-05,
-      "loss": 0.2605,
+      "loss": 0.2542,
       "step": 80
     },
     {
       "epoch": 1.8947368421052633,
-      "grad_norm": 6.902656078338623,
+      "grad_norm": 7.690194606781006,
       "learning_rate": 2.023809523809524e-05,
-      "loss": 0.2492,
+      "loss": 0.2514,
       "step": 90
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9455555555555556,
-      "eval_loss": 0.1588747352361679,
-      "eval_runtime": 6.4788,
-      "eval_samples_per_second": 416.741,
-      "eval_steps_per_second": 3.396,
+      "eval_accuracy": 0.9462962962962963,
+      "eval_loss": 0.16330811381340027,
+      "eval_runtime": 5.834,
+      "eval_samples_per_second": 462.801,
+      "eval_steps_per_second": 3.771,
       "step": 95
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 4.410822868347168,
+      "grad_norm": 6.186488628387451,
       "learning_rate": 1.626984126984127e-05,
-      "loss": 0.2391,
+      "loss": 0.2324,
       "step": 100
     },
     {
       "epoch": 2.3157894736842106,
-      "grad_norm": 6.006781101226807,
+      "grad_norm": 5.900567531585693,
       "learning_rate": 1.2301587301587301e-05,
-      "loss": 0.2186,
+      "loss": 0.2235,
       "step": 110
     },
     {
       "epoch": 2.526315789473684,
-      "grad_norm": 8.734150886535645,
+      "grad_norm": 6.816904544830322,
       "learning_rate": 8.333333333333334e-06,
       "loss": 0.2262,
       "step": 120
     },
     {
       "epoch": 2.736842105263158,
-      "grad_norm": 10.224825859069824,
+      "grad_norm": 5.449790000915527,
       "learning_rate": 4.365079365079365e-06,
-      "loss": 0.235,
+      "loss": 0.1988,
       "step": 130
     },
     {
       "epoch": 2.9473684210526314,
-      "grad_norm": 8.728509902954102,
+      "grad_norm": 7.056036472320557,
       "learning_rate": 3.9682539682539683e-07,
-      "loss": 0.2205,
+      "loss": 0.1907,
       "step": 140
     },
     {
       "epoch": 2.968421052631579,
-      "eval_accuracy": 0.9544444444444444,
-      "eval_loss": 0.13685034215450287,
-      "eval_runtime": 6.0237,
-      "eval_samples_per_second": 448.231,
-      "eval_steps_per_second": 3.652,
+      "eval_accuracy": 0.9507407407407408,
+      "eval_loss": 0.14795009791851044,
+      "eval_runtime": 6.5373,
+      "eval_samples_per_second": 413.015,
+      "eval_steps_per_second": 3.365,
       "step": 141
     },
     {
       "epoch": 2.968421052631579,
       "step": 141,
       "total_flos": 1.7938102147100836e+18,
-      "train_loss": 0.5156102181537777,
-      "train_runtime": 246.7274,
-      "train_samples_per_second": 295.468,
-      "train_steps_per_second": 0.571
+      "train_loss": 0.5108092158821458,
+      "train_runtime": 247.4957,
+      "train_samples_per_second": 294.551,
+      "train_steps_per_second": 0.57
     }
   ],
   "logging_steps": 10,