Training in progress, epoch 0

all_results.json

@@ -1,8 +1,13 @@
 {
     "epoch": 4.938271604938271,
+    "eval_accuracy": 0.9616724738675958,
+    "eval_loss": 0.11273417621850967,
+    "eval_runtime": 2.4706,
+    "eval_samples_per_second": 116.166,
+    "eval_steps_per_second": 7.286,
     "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.39835033178329465,
-    "train_runtime": 251.5945,
-    "train_samples_per_second": 51.333,
-    "train_steps_per_second": 0.795
+    "train_loss": 0.13397518873214723,
+    "train_runtime": 259.5354,
+    "train_samples_per_second": 49.762,
+    "train_steps_per_second": 0.771
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.938271604938271,
-    "eval_accuracy": 0.9547038327526133,
-    "eval_loss": 0.12142720818519592,
-    "eval_runtime": 2.5284,
-    "eval_samples_per_second": 113.511,
-    "eval_steps_per_second": 7.119
+    "eval_accuracy": 0.9616724738675958,
+    "eval_loss": 0.11273417621850967,
+    "eval_runtime": 2.4706,
+    "eval_samples_per_second": 116.166,
+    "eval_steps_per_second": 7.286
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ee704706002a80cbfd4afdb5adbaeb3b0c98b850ec8ccfab6cda940d54f0ab6b
+oid sha256:b0fc9da5c679d835c5b5188985e89e842b3b07dc57b274b60b5adc196921ba94
 size 110348984

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.938271604938271,
     "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.39835033178329465,
-    "train_runtime": 251.5945,
-    "train_samples_per_second": 51.333,
-    "train_steps_per_second": 0.795
+    "train_loss": 0.13397518873214723,
+    "train_runtime": 259.5354,
+    "train_samples_per_second": 49.762,
+    "train_steps_per_second": 0.771
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9442508710801394,
+  "best_metric": 0.9616724738675958,
   "best_model_checkpoint": "brain-tumor-classification/checkpoint-200",
   "epoch": 4.938271604938271,
   "eval_steps": 500,
@@ -10,92 +10,92 @@
   "log_history": [
     {
       "epoch": 0.9876543209876543,
-      "grad_norm": 26.882118225097656,
+      "grad_norm": 10.068439483642578,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.87,
+      "loss": 0.1888,
       "step": 40
     },
     {
       "epoch": 0.9876543209876543,
-      "eval_accuracy": 0.8710801393728222,
-      "eval_loss": 0.3874738812446594,
-      "eval_runtime": 4.2257,
-      "eval_samples_per_second": 67.917,
-      "eval_steps_per_second": 4.26,
+      "eval_accuracy": 0.9442508710801394,
+      "eval_loss": 0.15003159642219543,
+      "eval_runtime": 3.4689,
+      "eval_samples_per_second": 82.736,
+      "eval_steps_per_second": 5.189,
       "step": 40
     },
     {
       "epoch": 2.0,
-      "grad_norm": 6.630888938903809,
+      "grad_norm": 9.355758666992188,
       "learning_rate": 3.3055555555555553e-05,
-      "loss": 0.426,
+      "loss": 0.1514,
       "step": 81
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9024390243902439,
-      "eval_loss": 0.22352387011051178,
-      "eval_runtime": 2.3991,
-      "eval_samples_per_second": 119.628,
-      "eval_steps_per_second": 7.503,
+      "eval_accuracy": 0.9581881533101045,
+      "eval_loss": 0.12207956612110138,
+      "eval_runtime": 2.4115,
+      "eval_samples_per_second": 119.012,
+      "eval_steps_per_second": 7.464,
       "step": 81
     },
     {
       "epoch": 2.9876543209876543,
-      "grad_norm": 9.56891918182373,
+      "grad_norm": 11.253016471862793,
       "learning_rate": 2.1944444444444445e-05,
-      "loss": 0.292,
+      "loss": 0.1189,
       "step": 121
     },
     {
       "epoch": 2.9876543209876543,
-      "eval_accuracy": 0.926829268292683,
-      "eval_loss": 0.2056380957365036,
-      "eval_runtime": 2.3874,
-      "eval_samples_per_second": 120.214,
-      "eval_steps_per_second": 7.54,
+      "eval_accuracy": 0.9442508710801394,
+      "eval_loss": 0.16469331085681915,
+      "eval_runtime": 2.3749,
+      "eval_samples_per_second": 120.849,
+      "eval_steps_per_second": 7.579,
       "step": 121
     },
     {
       "epoch": 4.0,
-      "grad_norm": 7.695461273193359,
+      "grad_norm": 10.731207847595215,
       "learning_rate": 1.0555555555555555e-05,
-      "loss": 0.217,
+      "loss": 0.1042,
       "step": 162
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9337979094076655,
-      "eval_loss": 0.1578059196472168,
-      "eval_runtime": 2.6136,
-      "eval_samples_per_second": 109.811,
-      "eval_steps_per_second": 6.887,
+      "eval_accuracy": 0.9581881533101045,
+      "eval_loss": 0.12243915349245071,
+      "eval_runtime": 2.571,
+      "eval_samples_per_second": 111.632,
+      "eval_steps_per_second": 7.001,
       "step": 162
     },
     {
       "epoch": 4.938271604938271,
-      "grad_norm": 7.951013088226318,
+      "grad_norm": 7.372400283813477,
       "learning_rate": 0.0,
-      "loss": 0.1797,
+      "loss": 0.1055,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9442508710801394,
-      "eval_loss": 0.12732850015163422,
-      "eval_runtime": 2.9064,
-      "eval_samples_per_second": 98.748,
-      "eval_steps_per_second": 6.193,
+      "eval_accuracy": 0.9616724738675958,
+      "eval_loss": 0.11273417621850967,
+      "eval_runtime": 2.687,
+      "eval_samples_per_second": 106.809,
+      "eval_steps_per_second": 6.699,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
       "step": 200,
       "total_flos": 3.1727957353537536e+17,
-      "train_loss": 0.39835033178329465,
-      "train_runtime": 251.5945,
-      "train_samples_per_second": 51.333,
-      "train_steps_per_second": 0.795
+      "train_loss": 0.13397518873214723,
+      "train_runtime": 259.5354,
+      "train_samples_per_second": 49.762,
+      "train_steps_per_second": 0.771
     }
   ],
   "logging_steps": 10,