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": 0.15595091152609442,
-    "train_runtime": 2894.7808,
-    "train_samples_per_second": 3.15,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.151026108285837,
+    "train_runtime": 2890.8678,
+    "train_samples_per_second": 3.155,
+    "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e737cf3536910336e8c56ec31383ed9fd89c7121dab56c38c7155083049cd47f
+oid sha256:2371baab39a37cd98ece66a2f7efcc1edae6b53e867544735456cba5fd28e06a
 size 343223968

runs/Sep08_15-16-33_REDLGLAUCORE01/events.out.tfevents.1725836727.REDLGLAUCORE01.3667144.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a11e15ea8ce0dc8cc617685cdeb90152545f65d68c50fc94c7e98b9d1b72102a
+size 1057

runs/Sep09_14-14-49_REDLGLAUCORE01/events.out.tfevents.1725916490.REDLGLAUCORE01.3667144.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e9af543b1d300d1f600617a103b9d3087965844a8351280fef6afe5115b0f66
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.15595091152609442,
-    "train_runtime": 2894.7808,
-    "train_samples_per_second": 3.15,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.151026108285837,
+    "train_runtime": 2890.8678,
+    "train_samples_per_second": 3.155,
+    "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.11077258735895157,
+  "best_metric": 0.10812646150588989,
   "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": 1.4068071842193604,
+      "grad_norm": 1.534277319908142,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.5959,
+      "loss": 0.6017,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.9919264316558838,
+      "grad_norm": 1.6265959739685059,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4771,
+      "loss": 0.4577,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.2020492553710938,
+      "grad_norm": 1.137006163597107,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3854,
+      "loss": 0.3625,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.3166840076446533,
+      "grad_norm": 1.336227297782898,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3147,
+      "loss": 0.2812,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.7654367685317993,
+      "grad_norm": 1.6815255880355835,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2782,
+      "loss": 0.2571,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.8786159753799438,
+      "grad_norm": 0.7299286723136902,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.1948,
+      "loss": 0.1819,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.5508905649185181,
+      "grad_norm": 0.6613416075706482,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1734,
+      "loss": 0.175,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.6299371719360352,
+      "grad_norm": 0.6186805963516235,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1502,
+      "loss": 0.15,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 1.0293359756469727,
+      "grad_norm": 0.5209690928459167,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1214,
+      "loss": 0.1259,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 2.414964437484741,
+      "grad_norm": 1.0275017023086548,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1256,
+      "loss": 0.1277,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.9647577092511013,
+      "eval_accuracy": 0.9691629955947136,
       "eval_confusion_matrix": [
         [
-          58,
-          8
+          59,
+          7
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.1586856245994568,
-      "eval_runtime": 24.1166,
-      "eval_samples_per_second": 9.413,
-      "eval_steps_per_second": 0.332,
+      "eval_loss": 0.15413346886634827,
+      "eval_runtime": 25.0727,
+      "eval_samples_per_second": 9.054,
+      "eval_steps_per_second": 0.319,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 2.6475067138671875,
+      "grad_norm": 1.7517341375350952,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1269,
+      "loss": 0.1306,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.4995903968811035,
+      "grad_norm": 0.38637351989746094,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1145,
+      "loss": 0.1098,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.5841709971427917,
+      "grad_norm": 0.520092785358429,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1176,
+      "loss": 0.115,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.5116963982582092,
+      "grad_norm": 0.4652939736843109,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0822,
+      "loss": 0.0802,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.3481544852256775,
+      "grad_norm": 0.32277074456214905,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0919,
+      "loss": 0.0907,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.4802612066268921,
+      "grad_norm": 0.42583605647087097,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0846,
+      "loss": 0.0845,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 4.233730316162109,
+      "grad_norm": 0.45249637961387634,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0901,
+      "loss": 0.0824,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.34127503633499146,
+      "grad_norm": 0.32627880573272705,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0884,
+      "loss": 0.0894,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.3085730969905853,
+      "grad_norm": 0.33864378929138184,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0875,
+      "loss": 0.0804,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.939326822757721,
+      "grad_norm": 0.31026655435562134,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0888,
+      "loss": 0.0893,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.9691629955947136,
+      "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          62,
+          4
         ],
         [
           1,
           160
         ]
       ],
-      "eval_loss": 0.11077258735895157,
-      "eval_runtime": 24.8994,
-      "eval_samples_per_second": 9.117,
-      "eval_steps_per_second": 0.321,
+      "eval_loss": 0.10812646150588989,
+      "eval_runtime": 24.1908,
+      "eval_samples_per_second": 9.384,
+      "eval_steps_per_second": 0.331,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.3368864059448242,
+      "grad_norm": 0.2931233048439026,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1048,
+      "loss": 0.099,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.5311393737792969,
+      "grad_norm": 0.5454220175743103,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0682,
+      "loss": 0.0664,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.8554306030273438,
+      "grad_norm": 0.8330652117729187,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1024,
+      "loss": 0.1033,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.362326979637146,
+      "grad_norm": 0.35568487644195557,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0739,
+      "loss": 0.0746,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.3185746371746063,
+      "grad_norm": 0.28462159633636475,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0568,
+      "loss": 0.0573,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.2746303677558899,
+      "grad_norm": 0.2694823145866394,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.055,
+      "loss": 0.0536,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2818463146686554,
+      "grad_norm": 0.2652895152568817,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0477,
+      "loss": 0.0435,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.4329394996166229,
+      "grad_norm": 0.4277855157852173,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0999,
+      "loss": 0.0885,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.15595091152609442,
-      "train_runtime": 2894.7808,
-      "train_samples_per_second": 3.15,
-      "train_steps_per_second": 0.098
+      "train_loss": 0.151026108285837,
+      "train_runtime": 2890.8678,
+      "train_samples_per_second": 3.155,
+      "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:f50f07017ce1d77e3d381faa223eb14d9b12a6b9a138100ab36ae556a525c7c2
+oid sha256:46ba98f2464f87d2d3be52359f4843379c04d51706662ab0156f168acefe77b1
 size 5240