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": 0.16548631944154438,
-    "train_runtime": 2814.3096,
-    "train_samples_per_second": 3.241,
+    "train_loss": 0.15406067664163153,
+    "train_runtime": 2830.5602,
+    "train_samples_per_second": 3.222,
     "train_steps_per_second": 0.101
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8fcbaa0ea198acc4ea50a420fe0532f3483553fff85e94595521693ca29fc1d
+oid sha256:df3eb8066f3837014d78ea04775207f5e0e5d2f52a563d8568f5e8aba8b24ea4
 size 343223968

runs/Jul13_15-23-00_REDLGLAUCORE01/events.out.tfevents.1720912253.REDLGLAUCORE01.162992.34

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e47d90af6259543a604ded808d0190ec6cf37b2a60a315074fdbcee2bb76bc6d
+size 1380

runs/Jul19_12-56-15_REDLGLAUCORE01/events.out.tfevents.1721418976.REDLGLAUCORE01.323150.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aed197576633b2f480149a5d7acdd8fc78c0242211da16a87b611bfebd12676
+size 4184

runs/Jul19_12-58-59_REDLGLAUCORE01/events.out.tfevents.1721419139.REDLGLAUCORE01.543209.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f977e2ee6abf241960daad363e1527b5e82a7e3dfc647b5c1b4912031a432be8
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16548631944154438,
-    "train_runtime": 2814.3096,
-    "train_samples_per_second": 3.241,
+    "train_loss": 0.15406067664163153,
+    "train_runtime": 2830.5602,
+    "train_samples_per_second": 3.222,
     "train_steps_per_second": 0.101
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.12027937173843384,
+  "best_metric": 0.11733701080083847,
   "best_model_checkpoint": "/home/projects/Project - Fundus Photo Glaucoma Detection (AI)//model_results/vit-base-gpu/checkpoint-200",
   "epoch": 5.0,
   "eval_steps": 100,
@@ -10,72 +10,72 @@
   "log_history": [
     {
       "epoch": 0.17543859649122806,
-      "grad_norm": 0.9387953877449036,
+      "grad_norm": 1.1859416961669922,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6299,
+      "loss": 0.6066,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.4054632186889648,
+      "grad_norm": 1.6240363121032715,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.5138,
+      "loss": 0.4798,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.2556184530258179,
+      "grad_norm": 1.196411371231079,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.4223,
+      "loss": 0.3884,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.2750768661499023,
+      "grad_norm": 1.3352068662643433,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3304,
+      "loss": 0.312,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.3711668252944946,
+      "grad_norm": 1.4265506267547607,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.293,
+      "loss": 0.2724,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.8322893977165222,
+      "grad_norm": 0.7923513054847717,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2091,
+      "loss": 0.1841,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.8177331686019897,
+      "grad_norm": 0.6485021114349365,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.1938,
+      "loss": 0.1676,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 1.0095038414001465,
+      "grad_norm": 0.5704644322395325,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1586,
+      "loss": 0.1413,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 1.5976330041885376,
+      "grad_norm": 1.0736716985702515,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.134,
+      "loss": 0.117,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 3.159658908843994,
+      "grad_norm": 2.145616292953491,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1313,
+      "loss": 0.1265,
       "step": 100
     },
     {
@@ -91,164 +91,164 @@
           161
         ]
       ],
-      "eval_loss": 0.18016090989112854,
-      "eval_runtime": 25.5277,
-      "eval_samples_per_second": 8.892,
-      "eval_steps_per_second": 0.313,
+      "eval_loss": 0.16849496960639954,
+      "eval_runtime": 25.1005,
+      "eval_samples_per_second": 9.044,
+      "eval_steps_per_second": 0.319,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 4.4438323974609375,
+      "grad_norm": 5.284341335296631,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1357,
+      "loss": 0.1298,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.554610013961792,
+      "grad_norm": 0.568046510219574,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1229,
+      "loss": 0.1032,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.2898980677127838,
+      "grad_norm": 0.4078410863876343,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1219,
+      "loss": 0.1142,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.43265607953071594,
+      "grad_norm": 0.33205288648605347,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0867,
+      "loss": 0.0794,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.32516348361968994,
+      "grad_norm": 0.3029208183288574,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0936,
+      "loss": 0.0904,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.4287051558494568,
+      "grad_norm": 0.3194632828235626,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0874,
+      "loss": 0.0827,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 3.535736560821533,
+      "grad_norm": 2.0402252674102783,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0887,
+      "loss": 0.084,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3152014911174774,
+      "grad_norm": 0.3052498400211334,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0949,
+      "loss": 0.0868,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.28197869658470154,
+      "grad_norm": 0.28064411878585815,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0864,
+      "loss": 0.0822,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 4.776849269866943,
+      "grad_norm": 1.4272339344024658,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0978,
+      "loss": 0.091,
       "step": 200
     },
     {
       "epoch": 3.5087719298245617,
-      "eval_accuracy": 0.9691629955947136,
+      "eval_accuracy": 0.973568281938326,
       "eval_confusion_matrix": [
         [
-          59,
-          7
+          60,
+          6
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.12027937173843384,
-      "eval_runtime": 24.0716,
-      "eval_samples_per_second": 9.43,
-      "eval_steps_per_second": 0.332,
+      "eval_loss": 0.11733701080083847,
+      "eval_runtime": 24.7946,
+      "eval_samples_per_second": 9.155,
+      "eval_steps_per_second": 0.323,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.3222098648548126,
+      "grad_norm": 0.2653508484363556,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1081,
+      "loss": 0.1048,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.9184278845787048,
+      "grad_norm": 0.35800936818122864,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.074,
+      "loss": 0.0679,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.5896315574645996,
+      "grad_norm": 0.5756603479385376,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1126,
+      "loss": 0.105,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.3399348556995392,
+      "grad_norm": 0.2936393916606903,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0792,
+      "loss": 0.0762,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.26603880524635315,
+      "grad_norm": 0.25113916397094727,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0582,
+      "loss": 0.0554,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.25929567217826843,
+      "grad_norm": 0.2507961690425873,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0581,
+      "loss": 0.0546,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.26217120885849,
+      "grad_norm": 0.23755021393299103,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0473,
+      "loss": 0.0439,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.3724278509616852,
+      "grad_norm": 0.27338671684265137,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0989,
+      "loss": 0.0969,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16548631944154438,
-      "train_runtime": 2814.3096,
-      "train_samples_per_second": 3.241,
+      "train_loss": 0.15406067664163153,
+      "train_runtime": 2830.5602,
+      "train_samples_per_second": 3.222,
       "train_steps_per_second": 0.101
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13ec1fb1e758695cab098b0a771c394e325cbf8f4922a3674427573eeebfb4b1
+oid sha256:4fe7d7ebdefb05823bfd18e0fc815f9a7be5abc9c04fc742bf55224f4b0e55f5
 size 5240