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": 0.1531961118965818,
-    "train_runtime": 2875.9174,
-    "train_samples_per_second": 3.171,
+    "train_loss": 0.16874999728119164,
+    "train_runtime": 2873.7083,
+    "train_samples_per_second": 3.174,
     "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b6a9277b1148986c7dfb5d09a15f971af5eb7883556ac9edaecc3639da41eb5
+oid sha256:7101e7a2fa778f2d2e69a38c615bc94a0dd4fa5cb1ababb076e4c0358a8985da
 size 343223968

runs/Jul25_13-41-47_REDLGLAUCORE01/events.out.tfevents.1721943026.REDLGLAUCORE01.848854.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4d66ac4ece8e92009272d8011450dcee5e2c8a4da9e51f4a87aad1b6a3a558a
+size 1057

runs/Jul25_14-48-19_REDLGLAUCORE01/events.out.tfevents.1721944100.REDLGLAUCORE01.848854.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a1602057d12f6007a91b1a57fd98add85dd8e41598883b1e7fe132d7fb34f42
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.1531961118965818,
-    "train_runtime": 2875.9174,
-    "train_samples_per_second": 3.171,
+    "train_loss": 0.16874999728119164,
+    "train_runtime": 2873.7083,
+    "train_samples_per_second": 3.174,
     "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.10786318778991699,
+  "best_metric": 0.11347667127847672,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,245 +10,245 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 1.040859341621399,
+      "grad_norm": 1.0130871534347534,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6273,
+      "loss": 0.6237,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 2.026184320449829,
+      "grad_norm": 1.5061233043670654,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.5031,
+      "loss": 0.5133,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.257366418838501,
+      "grad_norm": 1.1895650625228882,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3756,
+      "loss": 0.4212,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.3905006647109985,
+      "grad_norm": 1.2210878133773804,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.2904,
+      "loss": 0.3366,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 0.6881418824195862,
+      "grad_norm": 0.8089302182197571,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2455,
+      "loss": 0.2959,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.7710828185081482,
+      "grad_norm": 0.8888680338859558,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.1673,
+      "loss": 0.2143,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.5303016304969788,
+      "grad_norm": 0.553264319896698,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1538,
+      "loss": 0.187,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.6288256049156189,
+      "grad_norm": 0.6764554381370544,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1481,
+      "loss": 0.161,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.5285879373550415,
+      "grad_norm": 0.9164976477622986,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1225,
+      "loss": 0.1341,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.38122934103012085,
+      "grad_norm": 0.4016127288341522,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1205,
+      "loss": 0.1269,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.9647577092511013,
+      "eval_accuracy": 0.9691629955947136,
       "eval_confusion_matrix": [
         [
-          58,
-          8
+          59,
+          7
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.15365372598171234,
-      "eval_runtime": 24.6174,
-      "eval_samples_per_second": 9.221,
-      "eval_steps_per_second": 0.325,
+      "eval_loss": 0.15991853177547455,
+      "eval_runtime": 24.3978,
+      "eval_samples_per_second": 9.304,
+      "eval_steps_per_second": 0.328,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 3.56727933883667,
+      "grad_norm": 2.8858230113983154,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1245,
+      "loss": 0.1424,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.666450023651123,
+      "grad_norm": 1.5411401987075806,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1098,
+      "loss": 0.121,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.4389030337333679,
+      "grad_norm": 0.32339081168174744,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.116,
+      "loss": 0.1263,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.3686705231666565,
+      "grad_norm": 0.4499027132987976,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0804,
+      "loss": 0.0902,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.33245834708213806,
+      "grad_norm": 0.403751939535141,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0902,
+      "loss": 0.0983,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.7682990431785583,
+      "grad_norm": 0.35326865315437317,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0842,
+      "loss": 0.091,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 1.1344654560089111,
+      "grad_norm": 1.8933607339859009,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0814,
+      "loss": 0.0917,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.29541823267936707,
+      "grad_norm": 0.34224092960357666,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0901,
+      "loss": 0.0996,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.3355918228626251,
+      "grad_norm": 0.38210582733154297,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0873,
+      "loss": 0.0967,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.591694176197052,
+      "grad_norm": 3.640868902206421,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0925,
+      "loss": 0.1017,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.973568281938326,
+      "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          61,
+          5
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.10786318778991699,
-      "eval_runtime": 25.607,
-      "eval_samples_per_second": 8.865,
-      "eval_steps_per_second": 0.312,
+      "eval_loss": 0.11347667127847672,
+      "eval_runtime": 22.6633,
+      "eval_samples_per_second": 10.016,
+      "eval_steps_per_second": 0.353,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.272243857383728,
+      "grad_norm": 0.3444685935974121,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.0988,
+      "loss": 0.1074,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.3665495216846466,
+      "grad_norm": 0.29822683334350586,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0696,
+      "loss": 0.0815,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.5697764158248901,
+      "grad_norm": 0.49087172746658325,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1108,
+      "loss": 0.1159,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.2947655916213989,
+      "grad_norm": 0.2734374701976776,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0753,
+      "loss": 0.0823,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.250372976064682,
+      "grad_norm": 0.30442044138908386,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0568,
+      "loss": 0.0662,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.24977755546569824,
+      "grad_norm": 0.30384954810142517,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.056,
+      "loss": 0.063,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2509959042072296,
+      "grad_norm": 0.3455353081226349,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0495,
+      "loss": 0.056,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.4848049581050873,
+      "grad_norm": 0.3426889479160309,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0922,
+      "loss": 0.1138,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.1531961118965818,
-      "train_runtime": 2875.9174,
-      "train_samples_per_second": 3.171,
+      "train_loss": 0.16874999728119164,
+      "train_runtime": 2873.7083,
+      "train_samples_per_second": 3.174,
       "train_steps_per_second": 0.099
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d91e7ec7fa23a915ae74c63046509bb159e4d88c5e41f8225ba75aacc01ea4d
+oid sha256:2c6fb497af13b66bd8a1f0d0ccf863bb8b087bd59e9c975e96a628fb51eb1cc1
 size 5240