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": 0.15237279649366411,
-    "train_runtime": 2884.1273,
-    "train_samples_per_second": 3.162,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.15217331428276865,
+    "train_runtime": 2935.6937,
+    "train_samples_per_second": 3.107,
+    "train_steps_per_second": 0.097
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:48514a3d3effb4dd526c6b1d4250ac9cfdd43e3bff313cb585ea9f297a33ca57
+oid sha256:f989f89f373a98e62c03e64c57d396fa3801c1c20193437fac071b607b251ef1
 size 343223968

runs/Sep06_16-45-57_REDLGLAUCORE01/events.out.tfevents.1725669337.REDLGLAUCORE01.848854.23

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24ac72044ddfee9b9edaaa20aa81c19b20e006444b66b6e29a7eb8c80d1cc020
+size 1057

runs/Sep07_18-44-43_REDLGLAUCORE01/events.out.tfevents.1725759884.REDLGLAUCORE01.3667144.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea0042134961ab00af551c6f9c4a27fed886c9b52e1b33ad079a165084ea75ff
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.15237279649366411,
-    "train_runtime": 2884.1273,
-    "train_samples_per_second": 3.162,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.15217331428276865,
+    "train_runtime": 2935.6937,
+    "train_samples_per_second": 3.107,
+    "train_steps_per_second": 0.097
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.10691466927528381,
+  "best_metric": 0.10852827876806259,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,70 +10,70 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 1.0408214330673218,
+      "grad_norm": 1.0400904417037964,
       "learning_rate": 9.649122807017545e-06,
       "loss": 0.6273,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 2.3301267623901367,
+      "grad_norm": 2.321244239807129,
       "learning_rate": 9.298245614035088e-06,
       "loss": 0.5016,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.2897900342941284,
+      "grad_norm": 1.2707430124282837,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3726,
+      "loss": 0.3722,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.2317367792129517,
+      "grad_norm": 1.238302230834961,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.2889,
+      "loss": 0.2875,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 0.6814886331558228,
+      "grad_norm": 0.7120734453201294,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2442,
+      "loss": 0.2431,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.7818458080291748,
+      "grad_norm": 0.7913864850997925,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.1655,
+      "loss": 0.1649,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.5145708322525024,
+      "grad_norm": 0.5288364887237549,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1518,
+      "loss": 0.1515,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.5460637807846069,
+      "grad_norm": 0.541506290435791,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1489,
+      "loss": 0.1479,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.5549706816673279,
+      "grad_norm": 0.5459899306297302,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1214,
+      "loss": 0.1204,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.3951248824596405,
+      "grad_norm": 0.38489124178886414,
       "learning_rate": 6.491228070175439e-06,
       "loss": 0.1191,
       "step": 100
@@ -91,165 +91,165 @@
           161
         ]
       ],
-      "eval_loss": 0.152234748005867,
-      "eval_runtime": 25.6232,
-      "eval_samples_per_second": 8.859,
-      "eval_steps_per_second": 0.312,
+      "eval_loss": 0.15112297236919403,
+      "eval_runtime": 25.4447,
+      "eval_samples_per_second": 8.921,
+      "eval_steps_per_second": 0.314,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 3.548973798751831,
+      "grad_norm": 3.6338634490966797,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1237,
+      "loss": 0.1244,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.8018013834953308,
+      "grad_norm": 0.5916204452514648,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.107,
+      "loss": 0.1078,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.44410237669944763,
+      "grad_norm": 0.43088626861572266,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1164,
+      "loss": 0.1147,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.38201433420181274,
+      "grad_norm": 0.355512410402298,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0798,
+      "loss": 0.0797,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.32631874084472656,
+      "grad_norm": 0.33613306283950806,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0899,
+      "loss": 0.0903,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.7354286313056946,
+      "grad_norm": 1.4403668642044067,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0839,
+      "loss": 0.084,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 0.7380229234695435,
+      "grad_norm": 1.5403963327407837,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0815,
+      "loss": 0.0825,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.2998797297477722,
+      "grad_norm": 0.3081526458263397,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0888,
+      "loss": 0.0893,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.324789434671402,
+      "grad_norm": 0.3143157958984375,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0868,
+      "loss": 0.0865,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.5307926535606384,
+      "grad_norm": 0.49979138374328613,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0899,
+      "loss": 0.09,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.9779735682819384,
+      "eval_accuracy": 0.973568281938326,
       "eval_confusion_matrix": [
         [
-          61,
-          5
+          60,
+          6
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.10691466927528381,
-      "eval_runtime": 24.2322,
-      "eval_samples_per_second": 9.368,
-      "eval_steps_per_second": 0.33,
+      "eval_loss": 0.10852827876806259,
+      "eval_runtime": 24.9381,
+      "eval_samples_per_second": 9.103,
+      "eval_steps_per_second": 0.321,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.2680317461490631,
+      "grad_norm": 0.26619821786880493,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.0987,
+      "loss": 0.0989,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.373812198638916,
+      "grad_norm": 0.3666665256023407,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0691,
+      "loss": 0.0687,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.6366022825241089,
+      "grad_norm": 0.7419259548187256,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1113,
+      "loss": 0.1107,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.3165888786315918,
+      "grad_norm": 0.3317352831363678,
       "learning_rate": 1.5789473684210526e-06,
       "loss": 0.075,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.2567109167575836,
+      "grad_norm": 0.27071481943130493,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0567,
+      "loss": 0.0571,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.24947808682918549,
+      "grad_norm": 0.2539863586425781,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0557,
+      "loss": 0.0559,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2490372359752655,
+      "grad_norm": 0.24817486107349396,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0488,
+      "loss": 0.0483,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.6002548336982727,
+      "grad_norm": 0.696506917476654,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0924,
+      "loss": 0.0915,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.15237279649366411,
-      "train_runtime": 2884.1273,
-      "train_samples_per_second": 3.162,
-      "train_steps_per_second": 0.099
+      "train_loss": 0.15217331428276865,
+      "train_runtime": 2935.6937,
+      "train_samples_per_second": 3.107,
+      "train_steps_per_second": 0.097
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d48392963a28e443df7f7049935e99f6dd751c4e47e1986efdc14051da4dcefa
+oid sha256:83b8cdb2e2ee45fcfa04f804448793e2c777d2b1e945f236f321e331655a93c2
 size 5240