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": 0.16674838442551462,
-    "train_runtime": 2899.7691,
-    "train_samples_per_second": 3.145,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.15243147653445863,
+    "train_runtime": 2870.7573,
+    "train_samples_per_second": 3.177,
+    "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9d92f743b75ab5c1ad7aafe4664965ff25823f047f832bc6648de1e166aa60b
+oid sha256:fe8e3ed1088ba67763a6ed0655f91bf86064e4bd36393229a8bcb39dcec6aa54
 size 343223968

runs/Oct05_15-11-08_REDLGLAUCORE01/events.out.tfevents.1728166269.REDLGLAUCORE01.299381.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75b040a9d52f2fb39e98a7740c0096280ff19554dca334b0a8eda1d8e9e235a5
+size 5257

runs/Oct05_15-11-59_REDLGLAUCORE01/events.out.tfevents.1728166319.REDLGLAUCORE01.299381.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cae93df799a9b8b9dd5be4da3325c9b993a0ce63bb8024e4335e37e548d2b4e5
+size 7437

runs/Sep24_13-38-28_REDLGLAUCORE01/events.out.tfevents.1727213226.REDLGLAUCORE01.5008.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0126f4d8c3783dc51043629b02094dbb98b668b47e9d32bb3be5baaad348ba1
+size 1057

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16674838442551462,
-    "train_runtime": 2899.7691,
-    "train_samples_per_second": 3.145,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.15243147653445863,
+    "train_runtime": 2870.7573,
+    "train_samples_per_second": 3.177,
+    "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.10996418446302414,
+  "best_metric": 0.11372088640928268,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,246 +10,246 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 1.0269614458084106,
+      "grad_norm": 1.413887858390808,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6502,
+      "loss": 0.5957,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.3762069940567017,
+      "grad_norm": 1.6277168989181519,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.5137,
+      "loss": 0.4704,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.176757574081421,
+      "grad_norm": 1.2440308332443237,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.4209,
+      "loss": 0.3788,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.234730839729309,
+      "grad_norm": 1.265623927116394,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3404,
+      "loss": 0.297,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.4434845447540283,
+      "grad_norm": 1.7784199714660645,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2967,
+      "loss": 0.2626,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.9210680723190308,
+      "grad_norm": 0.7343814373016357,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2193,
+      "loss": 0.1858,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.8182269334793091,
+      "grad_norm": 0.6488423347473145,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1973,
+      "loss": 0.171,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.683148980140686,
+      "grad_norm": 0.618161141872406,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1636,
+      "loss": 0.1463,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.7010196447372437,
+      "grad_norm": 0.5458269119262695,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1336,
+      "loss": 0.118,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.6333706974983215,
+      "grad_norm": 2.0569353103637695,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1306,
+      "loss": 0.1251,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.973568281938326,
+      "eval_accuracy": 0.9691629955947136,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          59,
+          7
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.15364386141300201,
-      "eval_runtime": 25.824,
-      "eval_samples_per_second": 8.79,
-      "eval_steps_per_second": 0.31,
+      "eval_loss": 0.15127457678318024,
+      "eval_runtime": 25.5355,
+      "eval_samples_per_second": 8.89,
+      "eval_steps_per_second": 0.313,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 4.199541091918945,
+      "grad_norm": 3.166415214538574,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1281,
+      "loss": 0.1271,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.45986390113830566,
+      "grad_norm": 0.40827080607414246,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1097,
+      "loss": 0.0974,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.45427531003952026,
+      "grad_norm": 0.3573634922504425,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1208,
+      "loss": 0.1152,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.47758427262306213,
+      "grad_norm": 0.5010245442390442,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0848,
+      "loss": 0.081,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.34909966588020325,
+      "grad_norm": 0.32326406240463257,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0956,
+      "loss": 0.0933,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.3922755718231201,
+      "grad_norm": 0.37870192527770996,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.088,
+      "loss": 0.0837,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 1.5563603639602661,
+      "grad_norm": 1.8692576885223389,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0902,
+      "loss": 0.0851,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3467804789543152,
+      "grad_norm": 0.3218236565589905,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0938,
+      "loss": 0.0895,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.40282759070396423,
+      "grad_norm": 0.284637987613678,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0915,
+      "loss": 0.0812,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.45585140585899353,
+      "grad_norm": 0.4369310438632965,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0912,
+      "loss": 0.0895,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.9779735682819384,
+      "eval_accuracy": 0.9691629955947136,
       "eval_confusion_matrix": [
         [
-          62,
-          4
+          60,
+          6
         ],
         [
           1,
           160
         ]
       ],
-      "eval_loss": 0.10996418446302414,
-      "eval_runtime": 25.5616,
-      "eval_samples_per_second": 8.88,
-      "eval_steps_per_second": 0.313,
+      "eval_loss": 0.11372088640928268,
+      "eval_runtime": 24.651,
+      "eval_samples_per_second": 9.209,
+      "eval_steps_per_second": 0.325,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.31239479780197144,
+      "grad_norm": 0.2749626934528351,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1078,
+      "loss": 0.1034,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.48598727583885193,
+      "grad_norm": 0.5386877059936523,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0727,
+      "loss": 0.0667,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.7427765130996704,
+      "grad_norm": 0.9905394315719604,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1106,
+      "loss": 0.1091,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.34395354986190796,
+      "grad_norm": 0.4203732907772064,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0804,
+      "loss": 0.0749,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.2963009178638458,
+      "grad_norm": 0.2867703139781952,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0615,
+      "loss": 0.0563,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.2988217771053314,
+      "grad_norm": 0.2705151438713074,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0589,
+      "loss": 0.0546,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2799382209777832,
+      "grad_norm": 0.2694481611251831,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0507,
+      "loss": 0.0432,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.3836091458797455,
+      "grad_norm": 0.3901878893375397,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.1024,
+      "loss": 0.096,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16674838442551462,
-      "train_runtime": 2899.7691,
-      "train_samples_per_second": 3.145,
-      "train_steps_per_second": 0.098
+      "train_loss": 0.15243147653445863,
+      "train_runtime": 2870.7573,
+      "train_samples_per_second": 3.177,
+      "train_steps_per_second": 0.099
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2eac6d503acb88f15eae4f2185e20bc2f9ce1308b51b9b46eead9d7f4c613c45
+oid sha256:73b80ca68f92aea3593ac4b70eeaff60c76c88359bdd51f9508f464ae5b83beb
 size 5240