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": 0.16555590545921994,
-    "train_runtime": 2847.4829,
-    "train_samples_per_second": 3.203,
-    "train_steps_per_second": 0.1
+    "train_loss": 0.16557477012015226,
+    "train_runtime": 2886.9237,
+    "train_samples_per_second": 3.159,
+    "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cde70ee2d0769eef51d1112bdd01931c56d2ff8e2591a00a2ef7f212c739ffa4
+oid sha256:9fb44775439f1568e346b080ef643007ca9d8cc2bd822693261c001502a04c6c
 size 343223968

runs/Sep18_11-12-19_REDLGLAUCORE01/events.out.tfevents.1726686074.REDLGLAUCORE01.3667144.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ade7b8572119237e6cd62a127a7f2181192c9bdf22c9cb38b9871671853fba2a
+size 1057

runs/Sep24_12-08-56_REDLGLAUCORE01/events.out.tfevents.1727204937.REDLGLAUCORE01.5008.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eeacf596c3a683ff264616a4c4970565d4f27b02c8a8b370d997e2d4fd79494d
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16555590545921994,
-    "train_runtime": 2847.4829,
-    "train_samples_per_second": 3.203,
-    "train_steps_per_second": 0.1
+    "train_loss": 0.16557477012015226,
+    "train_runtime": 2886.9237,
+    "train_samples_per_second": 3.159,
+    "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.1169881522655487,
+  "best_metric": 0.11615847796201706,
   "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": 1.276576280593872,
+      "grad_norm": 1.276593804359436,
       "learning_rate": 9.649122807017545e-06,
       "loss": 0.6232,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.7653061151504517,
+      "grad_norm": 1.7452043294906616,
       "learning_rate": 9.298245614035088e-06,
       "loss": 0.4728,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.360029697418213,
+      "grad_norm": 1.3645387887954712,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3828,
+      "loss": 0.383,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.4675012826919556,
+      "grad_norm": 1.6581600904464722,
       "learning_rate": 8.596491228070176e-06,
       "loss": 0.3044,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.1715320348739624,
+      "grad_norm": 1.1746941804885864,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2862,
+      "loss": 0.2863,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.816554844379425,
+      "grad_norm": 0.8200681805610657,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2171,
+      "loss": 0.2172,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.7754776477813721,
+      "grad_norm": 0.6782914996147156,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.2057,
+      "loss": 0.2049,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 1.184274435043335,
+      "grad_norm": 1.1690123081207275,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1824,
+      "loss": 0.182,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.7909060716629028,
+      "grad_norm": 0.7780042886734009,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.156,
+      "loss": 0.1556,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.45461058616638184,
+      "grad_norm": 0.4668606221675873,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1494,
+      "loss": 0.149,
       "step": 100
     },
     {
@@ -91,165 +91,165 @@
           161
         ]
       ],
-      "eval_loss": 0.17945599555969238,
-      "eval_runtime": 25.2577,
-      "eval_samples_per_second": 8.987,
-      "eval_steps_per_second": 0.317,
+      "eval_loss": 0.1784505844116211,
+      "eval_runtime": 25.5436,
+      "eval_samples_per_second": 8.887,
+      "eval_steps_per_second": 0.313,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 5.456608295440674,
+      "grad_norm": 5.776714324951172,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1443,
+      "loss": 0.1452,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.4778239130973816,
+      "grad_norm": 0.4670567810535431,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1275,
+      "loss": 0.1272,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.427962064743042,
+      "grad_norm": 0.4684557616710663,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1289,
+      "loss": 0.1283,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.49317681789398193,
+      "grad_norm": 0.49764347076416016,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0903,
+      "loss": 0.0891,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.36226317286491394,
+      "grad_norm": 0.36143022775650024,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.1,
+      "loss": 0.0999,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.4284095764160156,
+      "grad_norm": 0.4145030677318573,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0927,
+      "loss": 0.0926,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 2.2354979515075684,
+      "grad_norm": 2.2842023372650146,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0909,
+      "loss": 0.091,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.33503246307373047,
+      "grad_norm": 0.3328436613082886,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0981,
+      "loss": 0.0985,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.3614230155944824,
+      "grad_norm": 0.3577190339565277,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0889,
+      "loss": 0.0885,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.48562532663345337,
+      "grad_norm": 0.7198100686073303,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.1006,
+      "loss": 0.1007,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.973568281938326,
+      "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          61,
+          5
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.1169881522655487,
-      "eval_runtime": 25.5218,
-      "eval_samples_per_second": 8.894,
-      "eval_steps_per_second": 0.313,
+      "eval_loss": 0.11615847796201706,
+      "eval_runtime": 25.4517,
+      "eval_samples_per_second": 8.919,
+      "eval_steps_per_second": 0.314,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.38694825768470764,
+      "grad_norm": 0.39685574173927307,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.103,
+      "loss": 0.1031,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.4382840096950531,
+      "grad_norm": 0.557614803314209,
       "learning_rate": 2.280701754385965e-06,
       "loss": 0.0743,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.756301999092102,
+      "grad_norm": 0.762572169303894,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1059,
+      "loss": 0.1084,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.3622179925441742,
+      "grad_norm": 0.3623135983943939,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0779,
+      "loss": 0.0778,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.2922267019748688,
+      "grad_norm": 0.29174914956092834,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0614,
+      "loss": 0.0615,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.29758143424987793,
+      "grad_norm": 0.301498681306839,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0603,
+      "loss": 0.0601,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.28573909401893616,
+      "grad_norm": 0.2819487452507019,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0538,
+      "loss": 0.0539,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.556418240070343,
+      "grad_norm": 0.5504888892173767,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0902,
+      "loss": 0.0905,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16555590545921994,
-      "train_runtime": 2847.4829,
-      "train_samples_per_second": 3.203,
-      "train_steps_per_second": 0.1
+      "train_loss": 0.16557477012015226,
+      "train_runtime": 2886.9237,
+      "train_samples_per_second": 3.159,
+      "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 sha256:0a9cbfda4497da1135d2849b838bca022560bd69b7ee12748bbd45d4341ce28a
+oid sha256:bf812da5c442364f740187ca9b889e670e6468c2c5959ae8503a123337b24b73
 size 5240