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": 0.151026108285837,
-    "train_runtime": 2890.8678,
-    "train_samples_per_second": 3.155,
+    "train_loss": 0.1656212309996287,
+    "train_runtime": 2867.8524,
+    "train_samples_per_second": 3.18,
     "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:66ee3f61e20dc2ebbd33f11cf14028f514cbd8973a4f7919681cc6cc2f345f17
+oid sha256:2e2acdc32468a743f309b5b20ea99e483cb0f49cc850dfe933176088ace6d7b7
 size 343223968

runs/Sep09_14-14-49_REDLGLAUCORE01/events.out.tfevents.1725919401.REDLGLAUCORE01.3667144.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:231b64d85d5a9907e23ecd3dff07fa3d46e1c65fbbe7b979101536f6d6d1847f
+size 1057

runs/Sep11_08-54-13_REDLGLAUCORE01/events.out.tfevents.1726070053.REDLGLAUCORE01.3667144.10

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8da3602b1acd39c7eb5ed5df44da53de05eb11bcbb0f71a88778847a5a9aedf
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.151026108285837,
-    "train_runtime": 2890.8678,
-    "train_samples_per_second": 3.155,
+    "train_loss": 0.1656212309996287,
+    "train_runtime": 2867.8524,
+    "train_samples_per_second": 3.18,
     "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.10812646150588989,
+  "best_metric": 0.11660724878311157,
   "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": 1.534277319908142,
+      "grad_norm": 1.276576280593872,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6017,
+      "loss": 0.6232,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.6265959739685059,
+      "grad_norm": 1.7653061151504517,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4577,
+      "loss": 0.4728,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.137006163597107,
+      "grad_norm": 1.3600224256515503,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3625,
+      "loss": 0.3828,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.336227297782898,
+      "grad_norm": 1.4679501056671143,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.2812,
+      "loss": 0.3044,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.6815255880355835,
+      "grad_norm": 1.1724063158035278,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.2571,
+      "loss": 0.2862,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.7299286723136902,
+      "grad_norm": 0.8161798715591431,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.1819,
+      "loss": 0.217,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6613416075706482,
+      "grad_norm": 0.6667922735214233,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.175,
+      "loss": 0.2047,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 0.6186805963516235,
+      "grad_norm": 1.1511715650558472,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.15,
+      "loss": 0.1821,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.5209690928459167,
+      "grad_norm": 0.8299317359924316,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1259,
+      "loss": 0.1556,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 1.0275017023086548,
+      "grad_norm": 0.4476878046989441,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.1277,
+      "loss": 0.1493,
       "step": 100
     },
     {
       "epoch": 1.7543859649122808,
-      "eval_accuracy": 0.9691629955947136,
+      "eval_accuracy": 0.9647577092511013,
       "eval_confusion_matrix": [
         [
-          59,
-          7
+          58,
+          8
         ],
         [
           0,
           161
         ]
       ],
-      "eval_loss": 0.15413346886634827,
-      "eval_runtime": 25.0727,
-      "eval_samples_per_second": 9.054,
-      "eval_steps_per_second": 0.319,
+      "eval_loss": 0.1793680638074875,
+      "eval_runtime": 24.285,
+      "eval_samples_per_second": 9.347,
+      "eval_steps_per_second": 0.329,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 1.7517341375350952,
+      "grad_norm": 5.393967151641846,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1306,
+      "loss": 0.1456,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.38637351989746094,
+      "grad_norm": 0.469459593296051,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1098,
+      "loss": 0.1277,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.520092785358429,
+      "grad_norm": 0.42347291111946106,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.115,
+      "loss": 0.1297,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.4652939736843109,
+      "grad_norm": 0.490871787071228,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0802,
+      "loss": 0.0897,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.32277074456214905,
+      "grad_norm": 0.3595890402793884,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0907,
+      "loss": 0.0999,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.42583605647087097,
+      "grad_norm": 0.4167119562625885,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0845,
+      "loss": 0.0926,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 0.45249637961387634,
+      "grad_norm": 2.0740127563476562,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0824,
+      "loss": 0.0909,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.32627880573272705,
+      "grad_norm": 0.34823891520500183,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0894,
+      "loss": 0.0983,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.33864378929138184,
+      "grad_norm": 0.3568525016307831,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0804,
+      "loss": 0.0895,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.31026655435562134,
+      "grad_norm": 0.5340443849563599,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0893,
+      "loss": 0.1001,
       "step": 200
     },
     {
@@ -172,83 +172,83 @@
       "eval_accuracy": 0.9779735682819384,
       "eval_confusion_matrix": [
         [
-          62,
-          4
+          61,
+          5
         ],
         [
-          1,
-          160
+          0,
+          161
         ]
       ],
-      "eval_loss": 0.10812646150588989,
-      "eval_runtime": 24.1908,
-      "eval_samples_per_second": 9.384,
-      "eval_steps_per_second": 0.331,
+      "eval_loss": 0.11660724878311157,
+      "eval_runtime": 24.305,
+      "eval_samples_per_second": 9.34,
+      "eval_steps_per_second": 0.329,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.2931233048439026,
+      "grad_norm": 0.38332486152648926,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.099,
+      "loss": 0.103,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.5454220175743103,
+      "grad_norm": 0.661551296710968,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0664,
+      "loss": 0.0742,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.8330652117729187,
+      "grad_norm": 0.7470467686653137,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1033,
+      "loss": 0.1082,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.35568487644195557,
+      "grad_norm": 0.360360324382782,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0746,
+      "loss": 0.078,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.28462159633636475,
+      "grad_norm": 0.29448771476745605,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0573,
+      "loss": 0.0614,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.2694823145866394,
+      "grad_norm": 0.29900696873664856,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0536,
+      "loss": 0.0602,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.2652895152568817,
+      "grad_norm": 0.2832123339176178,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0435,
+      "loss": 0.0529,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.4277855157852173,
+      "grad_norm": 0.5533156991004944,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0885,
+      "loss": 0.0906,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.151026108285837,
-      "train_runtime": 2890.8678,
-      "train_samples_per_second": 3.155,
+      "train_loss": 0.1656212309996287,
+      "train_runtime": 2867.8524,
+      "train_samples_per_second": 3.18,
       "train_steps_per_second": 0.099
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46ba98f2464f87d2d3be52359f4843379c04d51706662ab0156f168acefe77b1
+oid sha256:73efdb6ee7ebf7b3727fc9398858b9ed1c22d8e58088a72231906727d5570500
 size 5240