Training in progress, epoch 0

all_results.json

@@ -6,8 +6,8 @@
     "eval_samples_per_second": 116.166,
     "eval_steps_per_second": 7.286,
     "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.13397518873214723,
-    "train_runtime": 259.5354,
-    "train_samples_per_second": 49.762,
-    "train_steps_per_second": 0.771
+    "train_loss": 0.06341908916831017,
+    "train_runtime": 245.9692,
+    "train_samples_per_second": 52.507,
+    "train_steps_per_second": 0.813
 }

config.json

@@ -59,7 +59,7 @@
     "stage4"
   ],
   "torch_dtype": "float32",
-  "transformers_version": "4.44.2",
+  "transformers_version": "4.46.0.dev0",
   "use_absolute_embeddings": false,
   "window_size": 7
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6dd27f60fc5f5bef62b6aaf60db59dfd490fd54f1bb9d7a9784ae221848217cb
+oid sha256:310684eab5eb0400be1adca0d599d477e0a680fe400c640a44e7216184a5bbcb
 size 110348984

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.938271604938271,
     "total_flos": 3.1727957353537536e+17,
-    "train_loss": 0.13397518873214723,
-    "train_runtime": 259.5354,
-    "train_samples_per_second": 49.762,
-    "train_steps_per_second": 0.771
+    "train_loss": 0.06341908916831017,
+    "train_runtime": 245.9692,
+    "train_samples_per_second": 52.507,
+    "train_steps_per_second": 0.813
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9616724738675958,
+  "best_metric": 0.9651567944250871,
   "best_model_checkpoint": "brain-tumor-classification/checkpoint-200",
   "epoch": 4.938271604938271,
   "eval_steps": 500,
@@ -8,94 +8,199 @@
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
+    {
+      "epoch": 0.24691358024691357,
+      "grad_norm": 9.583415985107422,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0798,
+      "step": 10
+    },
+    {
+      "epoch": 0.49382716049382713,
+      "grad_norm": 4.763514518737793,
+      "learning_rate": 5e-05,
+      "loss": 0.0516,
+      "step": 20
+    },
+    {
+      "epoch": 0.7407407407407407,
+      "grad_norm": 3.712924003601074,
+      "learning_rate": 4.722222222222222e-05,
+      "loss": 0.0708,
+      "step": 30
+    },
     {
       "epoch": 0.9876543209876543,
-      "grad_norm": 10.068439483642578,
+      "grad_norm": 3.5341458320617676,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.1888,
+      "loss": 0.0628,
       "step": 40
     },
     {
       "epoch": 0.9876543209876543,
-      "eval_accuracy": 0.9442508710801394,
-      "eval_loss": 0.15003159642219543,
-      "eval_runtime": 3.4689,
-      "eval_samples_per_second": 82.736,
-      "eval_steps_per_second": 5.189,
+      "eval_accuracy": 0.9512195121951219,
+      "eval_loss": 0.14935272932052612,
+      "eval_runtime": 2.3845,
+      "eval_samples_per_second": 120.36,
+      "eval_steps_per_second": 7.549,
       "step": 40
     },
     {
-      "epoch": 2.0,
-      "grad_norm": 9.355758666992188,
-      "learning_rate": 3.3055555555555553e-05,
-      "loss": 0.1514,
-      "step": 81
+      "epoch": 1.2345679012345678,
+      "grad_norm": 2.3918871879577637,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.0562,
+      "step": 50
+    },
+    {
+      "epoch": 1.4814814814814814,
+      "grad_norm": 7.054160118103027,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 0.0587,
+      "step": 60
+    },
+    {
+      "epoch": 1.7283950617283952,
+      "grad_norm": 16.58973503112793,
+      "learning_rate": 3.611111111111111e-05,
+      "loss": 0.0679,
+      "step": 70
+    },
+    {
+      "epoch": 1.9753086419753085,
+      "grad_norm": 11.970051765441895,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.0499,
+      "step": 80
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9581881533101045,
-      "eval_loss": 0.12207956612110138,
-      "eval_runtime": 2.4115,
-      "eval_samples_per_second": 119.012,
-      "eval_steps_per_second": 7.464,
+      "eval_accuracy": 0.9372822299651568,
+      "eval_loss": 0.18301770091056824,
+      "eval_runtime": 2.3848,
+      "eval_samples_per_second": 120.346,
+      "eval_steps_per_second": 7.548,
       "step": 81
     },
     {
-      "epoch": 2.9876543209876543,
-      "grad_norm": 11.253016471862793,
-      "learning_rate": 2.1944444444444445e-05,
-      "loss": 0.1189,
-      "step": 121
+      "epoch": 2.2222222222222223,
+      "grad_norm": 6.739134788513184,
+      "learning_rate": 3.055555555555556e-05,
+      "loss": 0.0352,
+      "step": 90
+    },
+    {
+      "epoch": 2.4691358024691357,
+      "grad_norm": 23.700721740722656,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.1,
+      "step": 100
+    },
+    {
+      "epoch": 2.7160493827160495,
+      "grad_norm": 6.7289581298828125,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0504,
+      "step": 110
+    },
+    {
+      "epoch": 2.962962962962963,
+      "grad_norm": 11.606633186340332,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.0697,
+      "step": 120
     },
     {
       "epoch": 2.9876543209876543,
-      "eval_accuracy": 0.9442508710801394,
-      "eval_loss": 0.16469331085681915,
-      "eval_runtime": 2.3749,
-      "eval_samples_per_second": 120.849,
-      "eval_steps_per_second": 7.579,
+      "eval_accuracy": 0.9547038327526133,
+      "eval_loss": 0.10937974601984024,
+      "eval_runtime": 2.9179,
+      "eval_samples_per_second": 98.36,
+      "eval_steps_per_second": 6.169,
       "step": 121
     },
     {
-      "epoch": 4.0,
-      "grad_norm": 10.731207847595215,
-      "learning_rate": 1.0555555555555555e-05,
-      "loss": 0.1042,
-      "step": 162
+      "epoch": 3.2098765432098766,
+      "grad_norm": 17.90212059020996,
+      "learning_rate": 1.9444444444444445e-05,
+      "loss": 0.0546,
+      "step": 130
+    },
+    {
+      "epoch": 3.45679012345679,
+      "grad_norm": 8.192621231079102,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.0601,
+      "step": 140
+    },
+    {
+      "epoch": 3.7037037037037037,
+      "grad_norm": 3.0072553157806396,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.07,
+      "step": 150
+    },
+    {
+      "epoch": 3.950617283950617,
+      "grad_norm": 5.344573974609375,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.0646,
+      "step": 160
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9581881533101045,
-      "eval_loss": 0.12243915349245071,
-      "eval_runtime": 2.571,
-      "eval_samples_per_second": 111.632,
-      "eval_steps_per_second": 7.001,
+      "eval_loss": 0.13176245987415314,
+      "eval_runtime": 2.8306,
+      "eval_samples_per_second": 101.391,
+      "eval_steps_per_second": 6.359,
       "step": 162
     },
+    {
+      "epoch": 4.197530864197531,
+      "grad_norm": 12.885249137878418,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0515,
+      "step": 170
+    },
+    {
+      "epoch": 4.444444444444445,
+      "grad_norm": 13.122868537902832,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 0.0784,
+      "step": 180
+    },
+    {
+      "epoch": 4.6913580246913575,
+      "grad_norm": 8.814871788024902,
+      "learning_rate": 2.777777777777778e-06,
+      "loss": 0.088,
+      "step": 190
+    },
     {
       "epoch": 4.938271604938271,
-      "grad_norm": 7.372400283813477,
+      "grad_norm": 1.9182705879211426,
       "learning_rate": 0.0,
-      "loss": 0.1055,
+      "loss": 0.0481,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
-      "eval_accuracy": 0.9616724738675958,
-      "eval_loss": 0.11273417621850967,
-      "eval_runtime": 2.687,
-      "eval_samples_per_second": 106.809,
-      "eval_steps_per_second": 6.699,
+      "eval_accuracy": 0.9651567944250871,
+      "eval_loss": 0.11668162047863007,
+      "eval_runtime": 2.6735,
+      "eval_samples_per_second": 107.351,
+      "eval_steps_per_second": 6.733,
       "step": 200
     },
     {
       "epoch": 4.938271604938271,
       "step": 200,
       "total_flos": 3.1727957353537536e+17,
-      "train_loss": 0.13397518873214723,
-      "train_runtime": 259.5354,
-      "train_samples_per_second": 49.762,
-      "train_steps_per_second": 0.771
+      "train_loss": 0.06341908916831017,
+      "train_runtime": 245.9692,
+      "train_samples_per_second": 52.507,
+      "train_steps_per_second": 0.813
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:131e591198da2fbb131a5c4950150736663b53e86d59e33c5e636c478d4cf168
-size 5176
+oid sha256:7ac1b4cd2d8159af1fc8fd51f45445f24aa9bf4ee3eaf6c19dbdcdb4ce9f1e54
+size 5240