Training in progress, step 100

all_results.json

@@ -5,8 +5,8 @@
     "eval_samples_per_second": 9.337,
     "eval_steps_per_second": 0.294,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.15408909237175658,
-    "train_runtime": 2777.0108,
-    "train_samples_per_second": 3.284,
-    "train_steps_per_second": 0.103
+    "train_loss": 0.16548631944154438,
+    "train_runtime": 2814.3096,
+    "train_samples_per_second": 3.241,
+    "train_steps_per_second": 0.101
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1bd29b5d92aad5541ff7d429d24052f6d6296d2be057219ae678d44ceeef32a
+oid sha256:57ea2893ff8e8b28de07577d434e93654585b172bd86ebd7429a0fb4aafba2f1
 size 343223968

runs/Jul13_14-20-53_REDLGLAUCORE01/events.out.tfevents.1720908509.REDLGLAUCORE01.162992.31

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97adbce150645e7d6c36c990a9116835945fea28c301fd057dc7273c71250978
+size 734

runs/Jul13_15-23-00_REDLGLAUCORE01/events.out.tfevents.1720909381.REDLGLAUCORE01.162992.33

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:288a75dc4cb5201a5209c15179e0870454bfc5bb34237cd4bc9cc67d23626a32
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.15408909237175658,
-    "train_runtime": 2777.0108,
-    "train_samples_per_second": 3.284,
-    "train_steps_per_second": 0.103
+    "train_loss": 0.16548631944154438,
+    "train_runtime": 2814.3096,
+    "train_samples_per_second": 3.241,
+    "train_steps_per_second": 0.101
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.11776807904243469,
+  "best_metric": 0.12027937173843384,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,72 +10,72 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 1.1859416961669922,
+      "grad_norm": 0.9387953877449036,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6066,
+      "loss": 0.6299,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.6240363121032715,
+      "grad_norm": 1.4054632186889648,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4798,
+      "loss": 0.5138,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.1964203119277954,
+      "grad_norm": 1.2556184530258179,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3884,
+      "loss": 0.4223,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.3346258401870728,
+      "grad_norm": 1.2750768661499023,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.312,
+      "loss": 0.3304,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.426996111869812,
+      "grad_norm": 1.3711668252944946,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2724,
+      "loss": 0.293,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.7832045555114746,
+      "grad_norm": 0.8322893977165222,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.1841,
+      "loss": 0.2091,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6473114490509033,
+      "grad_norm": 0.8177331686019897,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1676,
+      "loss": 0.1938,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.5747418999671936,
+      "grad_norm": 1.0095038414001465,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1413,
+      "loss": 0.1586,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 1.0683565139770508,
+      "grad_norm": 1.5976330041885376,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.117,
+      "loss": 0.134,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 2.0752248764038086,
+      "grad_norm": 3.159658908843994,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1265,
+      "loss": 0.1313,
       "step": 100
     },
     {
@@ -91,165 +91,165 @@
           161
         ]
       ],
-      "eval_loss": 0.16855977475643158,
-      "eval_runtime": 22.1189,
-      "eval_samples_per_second": 10.263,
-      "eval_steps_per_second": 0.362,
+      "eval_loss": 0.18016090989112854,
+      "eval_runtime": 25.5277,
+      "eval_samples_per_second": 8.892,
+      "eval_steps_per_second": 0.313,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 5.2471113204956055,
+      "grad_norm": 4.4438323974609375,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1295,
+      "loss": 0.1357,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.5815853476524353,
+      "grad_norm": 0.554610013961792,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1036,
+      "loss": 0.1229,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.4059981107711792,
+      "grad_norm": 0.2898980677127838,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1145,
+      "loss": 0.1219,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.3422783613204956,
+      "grad_norm": 0.43265607953071594,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0792,
+      "loss": 0.0867,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.3052961230278015,
+      "grad_norm": 0.32516348361968994,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0904,
+      "loss": 0.0936,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.3093198537826538,
+      "grad_norm": 0.4287051558494568,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0827,
+      "loss": 0.0874,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 2.0873372554779053,
+      "grad_norm": 3.535736560821533,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0841,
+      "loss": 0.0887,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3060620427131653,
+      "grad_norm": 0.3152014911174774,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0865,
+      "loss": 0.0949,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.28054070472717285,
+      "grad_norm": 0.28197869658470154,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0822,
+      "loss": 0.0864,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 1.5917067527770996,
+      "grad_norm": 4.776849269866943,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0911,
+      "loss": 0.0978,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.973568281938326,
+      "eval_accuracy": 0.9691629955947136,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          59,
+          7
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.11776807904243469,
-      "eval_runtime": 25.2222,
-      "eval_samples_per_second": 9.0,
-      "eval_steps_per_second": 0.317,
+      "eval_loss": 0.12027937173843384,
+      "eval_runtime": 24.0716,
+      "eval_samples_per_second": 9.43,
+      "eval_steps_per_second": 0.332,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.27064913511276245,
+      "grad_norm": 0.3222098648548126,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.105,
+      "loss": 0.1081,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.33931344747543335,
+      "grad_norm": 0.9184278845787048,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0679,
+      "loss": 0.074,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.5691994428634644,
+      "grad_norm": 0.5896315574645996,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1054,
+      "loss": 0.1126,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.29188117384910583,
+      "grad_norm": 0.3399348556995392,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0762,
+      "loss": 0.0792,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.24974815547466278,
+      "grad_norm": 0.26603880524635315,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0552,
+      "loss": 0.0582,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.25060197710990906,
+      "grad_norm": 0.25929567217826843,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0545,
+      "loss": 0.0581,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.237187922000885,
+      "grad_norm": 0.26217120885849,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0438,
+      "loss": 0.0473,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.27178001403808594,
+      "grad_norm": 0.3724278509616852,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0973,
+      "loss": 0.0989,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.15408909237175658,
-      "train_runtime": 2777.0108,
-      "train_samples_per_second": 3.284,
-      "train_steps_per_second": 0.103
+      "train_loss": 0.16548631944154438,
+      "train_runtime": 2814.3096,
+      "train_samples_per_second": 3.241,
+      "train_steps_per_second": 0.101
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1f857cead1585992cc7a2c3a4450d4cd4eee9b08fed6ed7241f61d2dd9629b8
+oid sha256:13ec1fb1e758695cab098b0a771c394e325cbf8f4922a3674427573eeebfb4b1
 size 5240