End of training

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 2.9655172413793105,
-    "eval_accuracy": 0.8990228013029316,
-    "eval_loss": 0.2956256568431854,
-    "eval_runtime": 16.6592,
-    "eval_samples_per_second": 18.428,
-    "eval_steps_per_second": 0.6,
+    "eval_accuracy": 0.9120521172638436,
+    "eval_loss": 0.22788342833518982,
+    "eval_runtime": 16.9379,
+    "eval_samples_per_second": 18.125,
+    "eval_steps_per_second": 0.59,
     "total_flos": 2.0203512366720614e+17,
-    "train_loss": 0.46530596036759636,
-    "train_runtime": 777.4672,
-    "train_samples_per_second": 10.638,
-    "train_steps_per_second": 0.081
+    "train_loss": 0.4531084828906589,
+    "train_runtime": 575.3091,
+    "train_samples_per_second": 14.377,
+    "train_steps_per_second": 0.11
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.9655172413793105,
-    "eval_accuracy": 0.8990228013029316,
-    "eval_loss": 0.2956256568431854,
-    "eval_runtime": 16.6592,
-    "eval_samples_per_second": 18.428,
-    "eval_steps_per_second": 0.6
+    "eval_accuracy": 0.9120521172638436,
+    "eval_loss": 0.22788342833518982,
+    "eval_runtime": 16.9379,
+    "eval_samples_per_second": 18.125,
+    "eval_steps_per_second": 0.59
 }

runs/Mar09_17-18-53_6ea84fd0780b/events.out.tfevents.1741542115.6ea84fd0780b.386.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4047bfde421b60c7492ea648d66840a218fdbb692478bf74412f4b9565cedde
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.9655172413793105,
     "total_flos": 2.0203512366720614e+17,
-    "train_loss": 0.46530596036759636,
-    "train_runtime": 777.4672,
-    "train_samples_per_second": 10.638,
-    "train_steps_per_second": 0.081
+    "train_loss": 0.4531084828906589,
+    "train_runtime": 575.3091,
+    "train_samples_per_second": 14.377,
+    "train_steps_per_second": 0.11
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.8990228013029316,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-42",
+  "best_metric": 0.9120521172638436,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-63",
   "epoch": 2.9655172413793105,
   "eval_steps": 500,
   "global_step": 63,
@@ -10,81 +10,81 @@
   "log_history": [
     {
       "epoch": 0.45977011494252873,
-      "grad_norm": 4.308095932006836,
+      "grad_norm": 5.491071701049805,
       "learning_rate": 4.732142857142857e-05,
-      "loss": 0.9696,
+      "loss": 1.0065,
       "step": 10
     },
     {
       "epoch": 0.9195402298850575,
-      "grad_norm": 19.417911529541016,
+      "grad_norm": 6.633999347686768,
       "learning_rate": 3.839285714285715e-05,
-      "loss": 0.5698,
+      "loss": 0.5843,
       "step": 20
     },
     {
       "epoch": 0.9655172413793104,
-      "eval_accuracy": 0.6644951140065146,
-      "eval_loss": 0.7485707402229309,
-      "eval_runtime": 18.0627,
-      "eval_samples_per_second": 16.996,
-      "eval_steps_per_second": 0.554,
+      "eval_accuracy": 0.8078175895765473,
+      "eval_loss": 0.44510841369628906,
+      "eval_runtime": 15.981,
+      "eval_samples_per_second": 19.21,
+      "eval_steps_per_second": 0.626,
       "step": 21
     },
     {
       "epoch": 1.4137931034482758,
-      "grad_norm": 11.978184700012207,
+      "grad_norm": 6.420283794403076,
       "learning_rate": 2.9464285714285718e-05,
-      "loss": 0.4488,
+      "loss": 0.3753,
       "step": 30
     },
     {
       "epoch": 1.8735632183908046,
-      "grad_norm": 5.141038417816162,
+      "grad_norm": 15.039899826049805,
       "learning_rate": 2.0535714285714285e-05,
-      "loss": 0.2988,
+      "loss": 0.3136,
       "step": 40
     },
     {
       "epoch": 1.9655172413793105,
-      "eval_accuracy": 0.8990228013029316,
-      "eval_loss": 0.2956256568431854,
-      "eval_runtime": 15.9558,
-      "eval_samples_per_second": 19.241,
-      "eval_steps_per_second": 0.627,
+      "eval_accuracy": 0.9022801302931596,
+      "eval_loss": 0.26217135787010193,
+      "eval_runtime": 16.1065,
+      "eval_samples_per_second": 19.061,
+      "eval_steps_per_second": 0.621,
       "step": 42
     },
     {
       "epoch": 2.367816091954023,
-      "grad_norm": 6.264738082885742,
+      "grad_norm": 3.5867278575897217,
       "learning_rate": 1.1607142857142857e-05,
-      "loss": 0.3026,
+      "loss": 0.2711,
       "step": 50
     },
     {
       "epoch": 2.8275862068965516,
-      "grad_norm": 5.892236232757568,
+      "grad_norm": 7.183843612670898,
       "learning_rate": 2.6785714285714285e-06,
-      "loss": 0.2695,
+      "loss": 0.2411,
       "step": 60
     },
     {
       "epoch": 2.9655172413793105,
-      "eval_accuracy": 0.8957654723127035,
-      "eval_loss": 0.259478896856308,
-      "eval_runtime": 16.6934,
-      "eval_samples_per_second": 18.39,
-      "eval_steps_per_second": 0.599,
+      "eval_accuracy": 0.9120521172638436,
+      "eval_loss": 0.22788342833518982,
+      "eval_runtime": 17.2089,
+      "eval_samples_per_second": 17.84,
+      "eval_steps_per_second": 0.581,
       "step": 63
     },
     {
       "epoch": 2.9655172413793105,
       "step": 63,
       "total_flos": 2.0203512366720614e+17,
-      "train_loss": 0.46530596036759636,
-      "train_runtime": 777.4672,
-      "train_samples_per_second": 10.638,
-      "train_steps_per_second": 0.081
+      "train_loss": 0.4531084828906589,
+      "train_runtime": 575.3091,
+      "train_samples_per_second": 14.377,
+      "train_steps_per_second": 0.11
     }
   ],
   "logging_steps": 10,