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": 0.16557477012015226,
-    "train_runtime": 2886.9237,
-    "train_samples_per_second": 3.159,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.16674838442551462,
+    "train_runtime": 2899.7691,
+    "train_samples_per_second": 3.145,
+    "train_steps_per_second": 0.098
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a2270d24ab316d6eb4c3d5476443386adc09833266cb4845e551dc5717c8d00
+oid sha256:1acf965e9a079384e7ba360a127a30402f358c19671435f280cf60ad282afa84
 size 343223968

runs/Sep24_12-08-56_REDLGLAUCORE01/events.out.tfevents.1727207889.REDLGLAUCORE01.5008.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81e6a423c77ba46555de0423e5965744f498bd48a4e41e6b1c33ff301b150a45
+size 1057

runs/Sep24_13-38-28_REDLGLAUCORE01/events.out.tfevents.1727210309.REDLGLAUCORE01.5008.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4830d839b92eb8f9ad860cae65a57dcbd6a35a3dc1d53719c98c876cd509bd89
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16557477012015226,
-    "train_runtime": 2886.9237,
-    "train_samples_per_second": 3.159,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.16674838442551462,
+    "train_runtime": 2899.7691,
+    "train_samples_per_second": 3.145,
+    "train_steps_per_second": 0.098
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.11615847796201706,
+  "best_metric": 0.10996418446302414,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,161 +10,161 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 1.276593804359436,
+      "grad_norm": 1.0269614458084106,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6232,
+      "loss": 0.6502,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.7452043294906616,
+      "grad_norm": 1.3762069940567017,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4728,
+      "loss": 0.5137,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.3645387887954712,
+      "grad_norm": 1.176757574081421,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.383,
+      "loss": 0.4209,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.6581600904464722,
+      "grad_norm": 1.234730839729309,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3044,
+      "loss": 0.3404,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.1746941804885864,
+      "grad_norm": 1.4434845447540283,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2863,
+      "loss": 0.2967,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.8200681805610657,
+      "grad_norm": 0.9210680723190308,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2172,
+      "loss": 0.2193,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6782914996147156,
+      "grad_norm": 0.8182269334793091,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.2049,
+      "loss": 0.1973,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 1.1690123081207275,
+      "grad_norm": 0.683148980140686,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.182,
+      "loss": 0.1636,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.7780042886734009,
+      "grad_norm": 0.7010196447372437,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1556,
+      "loss": 0.1336,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.4668606221675873,
+      "grad_norm": 0.6333706974983215,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.149,
+      "loss": 0.1306,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.9647577092511013,
+      "eval_accuracy": 0.973568281938326,
       "eval_confusion_matrix": [
         [
-          58,
-          8
+          60,
+          6
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.1784505844116211,
-      "eval_runtime": 25.5436,
-      "eval_samples_per_second": 8.887,
-      "eval_steps_per_second": 0.313,
+      "eval_loss": 0.15364386141300201,
+      "eval_runtime": 25.824,
+      "eval_samples_per_second": 8.79,
+      "eval_steps_per_second": 0.31,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 5.776714324951172,
+      "grad_norm": 4.199541091918945,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1452,
+      "loss": 0.1281,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.4670567810535431,
+      "grad_norm": 0.45986390113830566,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1272,
+      "loss": 0.1097,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.4684557616710663,
+      "grad_norm": 0.45427531003952026,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1283,
+      "loss": 0.1208,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.49764347076416016,
+      "grad_norm": 0.47758427262306213,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0891,
+      "loss": 0.0848,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.36143022775650024,
+      "grad_norm": 0.34909966588020325,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0999,
+      "loss": 0.0956,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.4145030677318573,
+      "grad_norm": 0.3922755718231201,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0926,
+      "loss": 0.088,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 2.2842023372650146,
+      "grad_norm": 1.5563603639602661,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.091,
+      "loss": 0.0902,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3328436613082886,
+      "grad_norm": 0.3467804789543152,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0985,
+      "loss": 0.0938,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.3577190339565277,
+      "grad_norm": 0.40282759070396423,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0885,
+      "loss": 0.0915,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.7198100686073303,
+      "grad_norm": 0.45585140585899353,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.1007,
+      "loss": 0.0912,
       "step": 200
     },
     {
@@ -172,84 +172,84 @@
       "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          61,
-          5
+          62,
+          4
         ],
         [
-          0,
-          161
+          1,
+          160
         ]
       ],
-      "eval_loss": 0.11615847796201706,
-      "eval_runtime": 25.4517,
-      "eval_samples_per_second": 8.919,
-      "eval_steps_per_second": 0.314,
+      "eval_loss": 0.10996418446302414,
+      "eval_runtime": 25.5616,
+      "eval_samples_per_second": 8.88,
+      "eval_steps_per_second": 0.313,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.39685574173927307,
+      "grad_norm": 0.31239479780197144,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1031,
+      "loss": 0.1078,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.557614803314209,
+      "grad_norm": 0.48598727583885193,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0743,
+      "loss": 0.0727,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.762572169303894,
+      "grad_norm": 0.7427765130996704,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1084,
+      "loss": 0.1106,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.3623135983943939,
+      "grad_norm": 0.34395354986190796,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0778,
+      "loss": 0.0804,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.29174914956092834,
+      "grad_norm": 0.2963009178638458,
       "learning_rate": 1.2280701754385965e-06,
       "loss": 0.0615,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.301498681306839,
+      "grad_norm": 0.2988217771053314,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0601,
+      "loss": 0.0589,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2819487452507019,
+      "grad_norm": 0.2799382209777832,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0539,
+      "loss": 0.0507,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.5504888892173767,
+      "grad_norm": 0.3836091458797455,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0905,
+      "loss": 0.1024,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16557477012015226,
-      "train_runtime": 2886.9237,
-      "train_samples_per_second": 3.159,
-      "train_steps_per_second": 0.099
+      "train_loss": 0.16674838442551462,
+      "train_runtime": 2899.7691,
+      "train_samples_per_second": 3.145,
+      "train_steps_per_second": 0.098
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf812da5c442364f740187ca9b889e670e6468c2c5959ae8503a123337b24b73
+oid sha256:2eac6d503acb88f15eae4f2185e20bc2f9ce1308b51b9b46eead9d7f4c613c45
 size 5240