End of training

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.9855072463768116,
     "total_flos": 1.686437022030889e+17,
-    "train_loss": 1.1870845416012932,
-    "train_runtime": 162.1308,
-    "train_samples_per_second": 13.538,
-    "train_steps_per_second": 0.105
+    "train_loss": 1.5057464066673727,
+    "train_runtime": 321.636,
+    "train_samples_per_second": 6.824,
+    "train_steps_per_second": 0.053
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.9855072463768116,
     "total_flos": 1.686437022030889e+17,
-    "train_loss": 1.1870845416012932,
-    "train_runtime": 162.1308,
-    "train_samples_per_second": 13.538,
-    "train_steps_per_second": 0.105
+    "train_loss": 1.5057464066673727,
+    "train_runtime": 321.636,
+    "train_samples_per_second": 6.824,
+    "train_steps_per_second": 0.053
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.3155737704918033,
+  "best_metric": 0.3483606557377049,
   "best_model_checkpoint": "doc-img-classification/checkpoint-17",
   "epoch": 0.9855072463768116,
   "eval_steps": 500,
@@ -10,44 +10,44 @@
   "log_history": [
     {
       "epoch": 0.057971014492753624,
-      "grad_norm": 3.7626097202301025,
+      "grad_norm": 7.618393898010254,
       "learning_rate": 0.0005,
-      "loss": 1.1898,
+      "loss": 1.2624,
       "step": 1
     },
     {
       "epoch": 0.6956521739130435,
-      "grad_norm": 2.0672760009765625,
+      "grad_norm": 7.588911056518555,
       "learning_rate": 0.0003333333333333333,
-      "loss": 1.2333,
+      "loss": 1.7064,
       "step": 12
     },
     {
       "epoch": 0.9855072463768116,
-      "eval_Macro F1": 0.2258236627984527,
-      "eval_Macro Precision": 0.24799453831711893,
-      "eval_Macro Recall": 0.3347186147186147,
-      "eval_Micro F1": 0.3155737704918033,
-      "eval_Micro Precision": 0.3155737704918033,
-      "eval_Micro Recall": 0.3155737704918033,
-      "eval_Weighted F1": 0.2118634338981494,
-      "eval_Weighted Precision": 0.231138727304777,
-      "eval_Weighted Recall": 0.3155737704918033,
-      "eval_accuracy": 0.3155737704918033,
-      "eval_loss": 1.099541425704956,
-      "eval_runtime": 14.6506,
-      "eval_samples_per_second": 16.655,
-      "eval_steps_per_second": 0.546,
+      "eval_Macro F1": 0.21733759318423854,
+      "eval_Macro Precision": 0.37635327635327637,
+      "eval_Macro Recall": 0.35452414439756214,
+      "eval_Micro F1": 0.3483606557377049,
+      "eval_Micro Precision": 0.3483606557377049,
+      "eval_Micro Recall": 0.3483606557377049,
+      "eval_Weighted F1": 0.21831561305190383,
+      "eval_Weighted Precision": 0.4016218299005184,
+      "eval_Weighted Recall": 0.3483606557377049,
+      "eval_accuracy": 0.3483606557377049,
+      "eval_loss": 1.0820339918136597,
+      "eval_runtime": 13.7157,
+      "eval_samples_per_second": 17.79,
+      "eval_steps_per_second": 0.583,
       "step": 17
     },
     {
       "epoch": 0.9855072463768116,
       "step": 17,
       "total_flos": 1.686437022030889e+17,
-      "train_loss": 1.1870845416012932,
-      "train_runtime": 162.1308,
-      "train_samples_per_second": 13.538,
-      "train_steps_per_second": 0.105
+      "train_loss": 1.5057464066673727,
+      "train_runtime": 321.636,
+      "train_samples_per_second": 6.824,
+      "train_steps_per_second": 0.053
     }
   ],
   "logging_steps": 12,