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": 0.16588184833526612,
-    "train_runtime": 2879.0892,
-    "train_samples_per_second": 3.168,
+    "train_loss": 0.15237279649366411,
+    "train_runtime": 2884.1273,
+    "train_samples_per_second": 3.162,
     "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b8d2bffc6f83501b18b62033f706d8e43919ef737ab3dd30ad516b8d817666e4
+oid sha256:56a1d9bc9dc168a7523e561e9e4528b5831e7c00df313213f5a0114cf54190fb
 size 343223968

runs/Sep05_20-01-33_REDLGLAUCORE01/events.out.tfevents.1725594629.REDLGLAUCORE01.848854.18

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61b530704c9bb5f33cf5751e0d7b944e58b5a567c55462d901d33af3107dd1d6
+size 1057

runs/Sep06_16-37-48_REDLGLAUCORE01/events.out.tfevents.1725665869.REDLGLAUCORE01.848854.20

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae8e9bd3c7ec7ce3b609ecb180d8d674dd065d7c4fe2395b0a2202276cf71ef0
+size 5464

runs/Sep06_16-45-57_REDLGLAUCORE01/events.out.tfevents.1725666357.REDLGLAUCORE01.848854.22

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bea66d5efd193cd5f160b9f9a3fa9f25b46e051841dc28d0d07d4fc30746cbaa
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16588184833526612,
-    "train_runtime": 2879.0892,
-    "train_samples_per_second": 3.168,
+    "train_loss": 0.15237279649366411,
+    "train_runtime": 2884.1273,
+    "train_samples_per_second": 3.162,
     "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.11670097708702087,
+  "best_metric": 0.10691466927528381,
   "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": 1.2752189636230469,
+      "grad_norm": 1.0408214330673218,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6232,
+      "loss": 0.6273,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.752753496170044,
+      "grad_norm": 2.3301267623901367,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.473,
+      "loss": 0.5016,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.3884831666946411,
+      "grad_norm": 1.2897900342941284,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3834,
+      "loss": 0.3726,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.6439697742462158,
+      "grad_norm": 1.2317367792129517,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.305,
+      "loss": 0.2889,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.1231735944747925,
+      "grad_norm": 0.6814886331558228,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2869,
+      "loss": 0.2442,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.8180190324783325,
+      "grad_norm": 0.7818458080291748,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2178,
+      "loss": 0.1655,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6753518581390381,
+      "grad_norm": 0.5145708322525024,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.206,
+      "loss": 0.1518,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 1.176467776298523,
+      "grad_norm": 0.5460637807846069,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1828,
+      "loss": 0.1489,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.7861670255661011,
+      "grad_norm": 0.5549706816673279,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1564,
+      "loss": 0.1214,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.4661114811897278,
+      "grad_norm": 0.3951248824596405,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1497,
+      "loss": 0.1191,
       "step": 100
     },
     {
@@ -91,80 +91,80 @@
           161
         ]
       ],
-      "eval_loss": 0.1807878464460373,
-      "eval_runtime": 24.5815,
-      "eval_samples_per_second": 9.235,
-      "eval_steps_per_second": 0.325,
+      "eval_loss": 0.152234748005867,
+      "eval_runtime": 25.6232,
+      "eval_samples_per_second": 8.859,
+      "eval_steps_per_second": 0.312,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 5.353541851043701,
+      "grad_norm": 3.548973798751831,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1463,
+      "loss": 0.1237,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.47052720189094543,
+      "grad_norm": 0.8018013834953308,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1291,
+      "loss": 0.107,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.43008866906166077,
+      "grad_norm": 0.44410237669944763,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1294,
+      "loss": 0.1164,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.495709627866745,
+      "grad_norm": 0.38201433420181274,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0904,
+      "loss": 0.0798,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.36288028955459595,
+      "grad_norm": 0.32631874084472656,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.1004,
+      "loss": 0.0899,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.42618927359580994,
+      "grad_norm": 0.7354286313056946,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0931,
+      "loss": 0.0839,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 1.7581294775009155,
+      "grad_norm": 0.7380229234695435,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0909,
+      "loss": 0.0815,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3336179554462433,
+      "grad_norm": 0.2998797297477722,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0991,
+      "loss": 0.0888,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.3549640476703644,
+      "grad_norm": 0.324789434671402,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.089,
+      "loss": 0.0868,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.6302672624588013,
+      "grad_norm": 0.5307926535606384,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0998,
+      "loss": 0.0899,
       "step": 200
     },
     {
@@ -180,75 +180,75 @@
           161
         ]
       ],
-      "eval_loss": 0.11670097708702087,
-      "eval_runtime": 25.3798,
-      "eval_samples_per_second": 8.944,
-      "eval_steps_per_second": 0.315,
+      "eval_loss": 0.10691466927528381,
+      "eval_runtime": 24.2322,
+      "eval_samples_per_second": 9.368,
+      "eval_steps_per_second": 0.33,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.3984730839729309,
+      "grad_norm": 0.2680317461490631,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1028,
+      "loss": 0.0987,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.4672948718070984,
+      "grad_norm": 0.373812198638916,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0746,
+      "loss": 0.0691,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.7369114756584167,
+      "grad_norm": 0.6366022825241089,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1065,
+      "loss": 0.1113,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.361775666475296,
+      "grad_norm": 0.3165888786315918,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0779,
+      "loss": 0.075,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.29275935888290405,
+      "grad_norm": 0.2567109167575836,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0615,
+      "loss": 0.0567,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.30065467953681946,
+      "grad_norm": 0.24947808682918549,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0603,
+      "loss": 0.0557,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.28486543893814087,
+      "grad_norm": 0.2490372359752655,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0523,
+      "loss": 0.0488,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.5634991526603699,
+      "grad_norm": 0.6002548336982727,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0905,
+      "loss": 0.0924,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16588184833526612,
-      "train_runtime": 2879.0892,
-      "train_samples_per_second": 3.168,
+      "train_loss": 0.15237279649366411,
+      "train_runtime": 2884.1273,
+      "train_samples_per_second": 3.162,
       "train_steps_per_second": 0.099
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:26ad7d3be1e9b00cda37081e9bd1f3b1066347be8b51b2fd04038001c72c8526
+oid sha256:d48392963a28e443df7f7049935e99f6dd751c4e47e1986efdc14051da4dcefa
 size 5240