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": 0.16532422492378637,
-    "train_runtime": 2902.1592,
-    "train_samples_per_second": 3.142,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.1531961118965818,
+    "train_runtime": 2875.9174,
+    "train_samples_per_second": 3.171,
+    "train_steps_per_second": 0.099
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abe2e52cb11b17a81e2dccc2b6f40ca0242bd66391d4ed5e288b5d3cb3b9f9b4
+oid sha256:27044ae21861e192102e42e9992dbd0ca1ac677f4d5e25fb259d962805ba57e0
 size 343223968

runs/Jul25_12-18-37_REDLGLAUCORE01/events.out.tfevents.1721938038.REDLGLAUCORE01.547362.52

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:214e196a5f7cb921305520b6a2b6e1f87948fd29e230a1837ec326d58a6e08bf
+size 1057

runs/Jul25_13-41-21_REDLGLAUCORE01/events.out.tfevents.1721940081.REDLGLAUCORE01.547362.54

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75d8e6bb931790eff68fb2304865f1ec000cc5b642a42d70abad28c21a254d54
+size 4184

runs/Jul25_13-41-47_REDLGLAUCORE01/events.out.tfevents.1721940107.REDLGLAUCORE01.848854.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a6c0b97513708a49ef1f12d474ea4bd8816c58d603aa5022e8450dd614c6632
+size 7437

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 7.020774259082035e+17,
-    "train_loss": 0.16532422492378637,
-    "train_runtime": 2902.1592,
-    "train_samples_per_second": 3.142,
-    "train_steps_per_second": 0.098
+    "train_loss": 0.1531961118965818,
+    "train_runtime": 2875.9174,
+    "train_samples_per_second": 3.171,
+    "train_steps_per_second": 0.099
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.11641146242618561,
+  "best_metric": 0.10786318778991699,
   "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": 1.276576280593872,
+      "grad_norm": 1.040859341621399,
       "learning_rate": 9.649122807017545e-06,
-      "loss": 0.6232,
+      "loss": 0.6273,
       "step": 10
     },
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 1.7671723365783691,
+      "grad_norm": 2.026184320449829,
       "learning_rate": 9.298245614035088e-06,
-      "loss": 0.4728,
+      "loss": 0.5031,
       "step": 20
     },
     {
       "epoch": 0.5263157894736842,
-      "grad_norm": 1.358464241027832,
+      "grad_norm": 1.257366418838501,
       "learning_rate": 8.947368421052632e-06,
-      "loss": 0.3828,
+      "loss": 0.3756,
       "step": 30
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 1.4662245512008667,
+      "grad_norm": 1.3905006647109985,
       "learning_rate": 8.596491228070176e-06,
-      "loss": 0.3042,
+      "loss": 0.2904,
       "step": 40
     },
     {
       "epoch": 0.8771929824561403,
-      "grad_norm": 1.1634291410446167,
+      "grad_norm": 0.6881418824195862,
       "learning_rate": 8.24561403508772e-06,
-      "loss": 0.286,
+      "loss": 0.2455,
       "step": 50
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 0.8176971077919006,
+      "grad_norm": 0.7710828185081482,
       "learning_rate": 7.894736842105265e-06,
-      "loss": 0.2169,
+      "loss": 0.1673,
       "step": 60
     },
     {
       "epoch": 1.2280701754385965,
-      "grad_norm": 0.6673957705497742,
+      "grad_norm": 0.5303016304969788,
       "learning_rate": 7.5438596491228074e-06,
-      "loss": 0.2044,
+      "loss": 0.1538,
       "step": 70
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 1.1810154914855957,
+      "grad_norm": 0.6288256049156189,
       "learning_rate": 7.192982456140352e-06,
-      "loss": 0.1821,
+      "loss": 0.1481,
       "step": 80
     },
     {
       "epoch": 1.5789473684210527,
-      "grad_norm": 0.759559690952301,
+      "grad_norm": 0.5285879373550415,
       "learning_rate": 6.842105263157896e-06,
-      "loss": 0.1558,
+      "loss": 0.1225,
       "step": 90
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 0.4498167037963867,
+      "grad_norm": 0.38122934103012085,
       "learning_rate": 6.491228070175439e-06,
-      "loss": 0.149,
+      "loss": 0.1205,
       "step": 100
     },
     {
@@ -91,80 +91,80 @@
           161
         ]
       ],
-      "eval_loss": 0.17813891172409058,
-      "eval_runtime": 25.2489,
-      "eval_samples_per_second": 8.99,
-      "eval_steps_per_second": 0.317,
+      "eval_loss": 0.15365372598171234,
+      "eval_runtime": 24.6174,
+      "eval_samples_per_second": 9.221,
+      "eval_steps_per_second": 0.325,
       "step": 100
     },
     {
       "epoch": 1.9298245614035088,
-      "grad_norm": 4.42154598236084,
+      "grad_norm": 3.56727933883667,
       "learning_rate": 6.140350877192983e-06,
-      "loss": 0.1435,
+      "loss": 0.1245,
       "step": 110
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 0.48428404331207275,
+      "grad_norm": 0.666450023651123,
       "learning_rate": 5.789473684210527e-06,
-      "loss": 0.1275,
+      "loss": 0.1098,
       "step": 120
     },
     {
       "epoch": 2.280701754385965,
-      "grad_norm": 0.41149669885635376,
+      "grad_norm": 0.4389030337333679,
       "learning_rate": 5.438596491228071e-06,
-      "loss": 0.1291,
+      "loss": 0.116,
       "step": 130
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 0.49232661724090576,
+      "grad_norm": 0.3686705231666565,
       "learning_rate": 5.087719298245615e-06,
-      "loss": 0.0896,
+      "loss": 0.0804,
       "step": 140
     },
     {
       "epoch": 2.6315789473684212,
-      "grad_norm": 0.36095741391181946,
+      "grad_norm": 0.33245834708213806,
       "learning_rate": 4.736842105263158e-06,
-      "loss": 0.0997,
+      "loss": 0.0902,
       "step": 150
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 0.4267736077308655,
+      "grad_norm": 0.7682990431785583,
       "learning_rate": 4.385964912280702e-06,
-      "loss": 0.0925,
+      "loss": 0.0842,
       "step": 160
     },
     {
       "epoch": 2.982456140350877,
-      "grad_norm": 1.9955291748046875,
+      "grad_norm": 1.1344654560089111,
       "learning_rate": 4.035087719298246e-06,
-      "loss": 0.0907,
+      "loss": 0.0814,
       "step": 170
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 0.3319329619407654,
+      "grad_norm": 0.29541823267936707,
       "learning_rate": 3.6842105263157896e-06,
-      "loss": 0.0979,
+      "loss": 0.0901,
       "step": 180
     },
     {
       "epoch": 3.3333333333333335,
-      "grad_norm": 0.35440099239349365,
+      "grad_norm": 0.3355918228626251,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.0886,
+      "loss": 0.0873,
       "step": 190
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 0.5289174318313599,
+      "grad_norm": 0.591694176197052,
       "learning_rate": 2.9824561403508774e-06,
-      "loss": 0.0996,
+      "loss": 0.0925,
       "step": 200
     },
     {
@@ -180,76 +180,76 @@
           161
         ]
       ],
-      "eval_loss": 0.11641146242618561,
-      "eval_runtime": 25.5874,
-      "eval_samples_per_second": 8.872,
-      "eval_steps_per_second": 0.313,
+      "eval_loss": 0.10786318778991699,
+      "eval_runtime": 25.607,
+      "eval_samples_per_second": 8.865,
+      "eval_steps_per_second": 0.312,
       "step": 200
     },
     {
       "epoch": 3.6842105263157894,
-      "grad_norm": 0.3807942569255829,
+      "grad_norm": 0.272243857383728,
       "learning_rate": 2.631578947368421e-06,
-      "loss": 0.1028,
+      "loss": 0.0988,
       "step": 210
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 0.6208366751670837,
+      "grad_norm": 0.3665495216846466,
       "learning_rate": 2.280701754385965e-06,
-      "loss": 0.0741,
+      "loss": 0.0696,
       "step": 220
     },
     {
       "epoch": 4.035087719298246,
-      "grad_norm": 0.7710980176925659,
+      "grad_norm": 0.5697764158248901,
       "learning_rate": 1.929824561403509e-06,
-      "loss": 0.1073,
+      "loss": 0.1108,
       "step": 230
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 0.36691102385520935,
+      "grad_norm": 0.2947655916213989,
       "learning_rate": 1.5789473684210526e-06,
-      "loss": 0.0777,
+      "loss": 0.0753,
       "step": 240
     },
     {
       "epoch": 4.385964912280702,
-      "grad_norm": 0.29168033599853516,
+      "grad_norm": 0.250372976064682,
       "learning_rate": 1.2280701754385965e-06,
-      "loss": 0.0614,
+      "loss": 0.0568,
       "step": 250
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 0.2981802821159363,
+      "grad_norm": 0.24977755546569824,
       "learning_rate": 8.771929824561404e-07,
-      "loss": 0.0601,
+      "loss": 0.056,
       "step": 260
     },
     {
       "epoch": 4.7368421052631575,
-      "grad_norm": 0.28405195474624634,
+      "grad_norm": 0.2509959042072296,
       "learning_rate": 5.263157894736843e-07,
-      "loss": 0.0525,
+      "loss": 0.0495,
       "step": 270
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 0.5508224964141846,
+      "grad_norm": 0.4848049581050873,
       "learning_rate": 1.7543859649122808e-07,
-      "loss": 0.0904,
+      "loss": 0.0922,
       "step": 280
     },
     {
       "epoch": 5.0,
       "step": 285,
       "total_flos": 7.020774259082035e+17,
-      "train_loss": 0.16532422492378637,
-      "train_runtime": 2902.1592,
-      "train_samples_per_second": 3.142,
-      "train_steps_per_second": 0.098
+      "train_loss": 0.1531961118965818,
+      "train_runtime": 2875.9174,
+      "train_samples_per_second": 3.171,
+      "train_steps_per_second": 0.099
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9bf02dbf635dd7132d5414a7b6df1609bc0998eeaa814356e850cb9332a72661
+oid sha256:2d91e7ec7fa23a915ae74c63046509bb159e4d88c5e41f8225ba75aacc01ea4d
 size 5240