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": 0.15312039318837617,
-    "train_runtime": 2880.1419,
-    "train_samples_per_second": 3.167,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.1566001436166596,
+    "train_runtime": 2850.1536,
+    "train_samples_per_second": 3.2,
+    "train_steps_per_second": 0.1
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2782d9af0de45298715f0323fbbd54ddf094bcb049aa280bdae42770c8f179d7
+oid sha256:2f9684d3aaa9461a75bc9fe057fc023069628ed3c75c60dc0968975fee82a463
 size 343223968

runs/Jul22_16-30-07_REDLGLAUCORE01/events.out.tfevents.1721693900.REDLGLAUCORE01.547362.18

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14b4d13e43b06666da438120675ebbffb77401158da128628d8bb025f11ce9f8
+size 1380

runs/Jul23_12-57-29_REDLGLAUCORE01/events.out.tfevents.1721764649.REDLGLAUCORE01.547362.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41e4dba0e0caa4ce43eb6d6b4b75822e3872f7b703e54ba3d294f38c98773f95
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.15312039318837617,
-    "train_runtime": 2880.1419,
-    "train_samples_per_second": 3.167,
-    "train_steps_per_second": 0.099
+    "train_loss": 0.1566001436166596,
+    "train_runtime": 2850.1536,
+    "train_samples_per_second": 3.2,
+    "train_steps_per_second": 0.1
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.1123642697930336,
+  "best_metric": 0.10913512110710144,
   "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": 1.4138169288635254,
+      "grad_norm": 0.9124495983123779,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.5957,
+      "loss": 0.6422,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.6343547105789185,
+      "grad_norm": 1.400277853012085,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4704,
+      "loss": 0.5107,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.1984628438949585,
+      "grad_norm": 1.2670001983642578,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3792,
+      "loss": 0.4099,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.2228361368179321,
+      "grad_norm": 1.2379003763198853,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3013,
+      "loss": 0.3261,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.6350319385528564,
+      "grad_norm": 1.3035060167312622,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.267,
+      "loss": 0.278,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.7450676560401917,
+      "grad_norm": 0.7487425804138184,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.189,
+      "loss": 0.1874,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6679130792617798,
+      "grad_norm": 1.2215185165405273,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1731,
+      "loss": 0.1672,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.6309726238250732,
+      "grad_norm": 0.8201051950454712,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1479,
+      "loss": 0.1454,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.5439296960830688,
+      "grad_norm": 1.6430377960205078,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1194,
+      "loss": 0.1211,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 2.2208399772644043,
+      "grad_norm": 3.366523504257202,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1265,
+      "loss": 0.124,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.9691629955947136,
+      "eval_accuracy": 0.9647577092511013,
       "eval_confusion_matrix": [
         [
-          59,
-          7
+          58,
+          8
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.15117686986923218,
-      "eval_runtime": 24.8749,
-      "eval_samples_per_second": 9.126,
-      "eval_steps_per_second": 0.322,
+      "eval_loss": 0.16817055642604828,
+      "eval_runtime": 25.0148,
+      "eval_samples_per_second": 9.075,
+      "eval_steps_per_second": 0.32,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 4.325542449951172,
+      "grad_norm": 5.223181247711182,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1281,
+      "loss": 0.1256,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.43080058693885803,
+      "grad_norm": 0.456160306930542,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.0971,
+      "loss": 0.1073,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.3750275671482086,
+      "grad_norm": 0.40346741676330566,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1155,
+      "loss": 0.1071,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.42711108922958374,
+      "grad_norm": 0.3802075684070587,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0805,
+      "loss": 0.0769,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.3187861740589142,
+      "grad_norm": 0.29936113953590393,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0936,
+      "loss": 0.0904,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.36738303303718567,
+      "grad_norm": 0.5365840792655945,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.084,
+      "loss": 0.0815,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 2.0587196350097656,
+      "grad_norm": 3.9614758491516113,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0855,
+      "loss": 0.0836,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.31759971380233765,
+      "grad_norm": 0.2877553701400757,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0896,
+      "loss": 0.086,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.2914780378341675,
+      "grad_norm": 0.27285733819007874,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0814,
+      "loss": 0.0808,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.33958643674850464,
+      "grad_norm": 1.5225030183792114,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0892,
+      "loss": 0.0881,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.9691629955947136,
+      "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          60,
-          6
+          61,
+          5
         ],
         [
-          1,
-          160
+          0,
+          161
         ]
       ],
-      "eval_loss": 0.1123642697930336,
-      "eval_runtime": 24.2578,
-      "eval_samples_per_second": 9.358,
-      "eval_steps_per_second": 0.33,
+      "eval_loss": 0.10913512110710144,
+      "eval_runtime": 24.033,
+      "eval_samples_per_second": 9.445,
+      "eval_steps_per_second": 0.333,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.27888378500938416,
+      "grad_norm": 0.2580512762069702,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1024,
+      "loss": 0.1032,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.526310920715332,
+      "grad_norm": 0.5844484567642212,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.067,
+      "loss": 0.0662,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 1.0086055994033813,
+      "grad_norm": 0.8993027210235596,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1094,
+      "loss": 0.0954,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.41842520236968994,
+      "grad_norm": 0.35451096296310425,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0748,
+      "loss": 0.0752,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.2806910574436188,
+      "grad_norm": 0.27109700441360474,
       "learning_rate": 1.2280701754385965e-06,
       "loss": 0.0563,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.2728028893470764,
+      "grad_norm": 0.24192890524864197,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.055,
+      "loss": 0.0535,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2732146978378296,
+      "grad_norm": 0.2452133148908615,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0434,
+      "loss": 0.0405,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.3774060010910034,
+      "grad_norm": 0.4464854598045349,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0962,
+      "loss": 0.0869,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.15312039318837617,
-      "train_runtime": 2880.1419,
-      "train_samples_per_second": 3.167,
-      "train_steps_per_second": 0.099
+      "train_loss": 0.1566001436166596,
+      "train_runtime": 2850.1536,
+      "train_samples_per_second": 3.2,
+      "train_steps_per_second": 0.1
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9b8fd544b0c9366b33522c4494fe94f39a7b2215a13ef22c15b438c785661593
+oid sha256:9a918e9d2444d2d031c19a1c82f4000ad8b839b20b58e9435864a7279430830b
 size 5240