End of training

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9120879120879121
+      value: 0.9194139194139194
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [facebook/vit-msn-small](https://huggingface.co/facebook/vit-msn-small) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2491
-- Accuracy: 0.9121
+- Loss: 0.2229
+- Accuracy: 0.9194
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 18.46153846153846,
-    "eval_accuracy": 0.9497584541062802,
-    "eval_loss": 0.1871330887079239,
-    "eval_runtime": 4.3162,
-    "eval_samples_per_second": 239.797,
-    "eval_steps_per_second": 3.939,
+    "eval_accuracy": 0.9194139194139194,
+    "eval_loss": 0.22294031083583832,
+    "eval_runtime": 1.2677,
+    "eval_samples_per_second": 215.358,
+    "eval_steps_per_second": 3.944,
     "total_flos": 2.9458314315627725e+17,
-    "train_loss": 0.35900469223658243,
-    "train_runtime": 212.0852,
-    "train_samples_per_second": 76.856,
-    "train_steps_per_second": 0.283
+    "train_loss": 0.342086935043335,
+    "train_runtime": 153.4118,
+    "train_samples_per_second": 106.25,
+    "train_steps_per_second": 0.391
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 18.46153846153846,
-    "eval_accuracy": 0.9497584541062802,
-    "eval_loss": 0.1871330887079239,
-    "eval_runtime": 4.3162,
-    "eval_samples_per_second": 239.797,
-    "eval_steps_per_second": 3.939
+    "eval_accuracy": 0.9194139194139194,
+    "eval_loss": 0.22294031083583832,
+    "eval_runtime": 1.2677,
+    "eval_samples_per_second": 215.358,
+    "eval_steps_per_second": 3.944
 }

runs/Jan14_17-44-41_c583982b4f3d/events.out.tfevents.1736876900.c583982b4f3d.215.9

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5e372d0453ced4d5ac7d7fcee3f40ec8bd5be65095c31c693c1a65c13d94143
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 18.46153846153846,
     "total_flos": 2.9458314315627725e+17,
-    "train_loss": 0.35900469223658243,
-    "train_runtime": 212.0852,
-    "train_samples_per_second": 76.856,
-    "train_steps_per_second": 0.283
+    "train_loss": 0.342086935043335,
+    "train_runtime": 153.4118,
+    "train_samples_per_second": 106.25,
+    "train_steps_per_second": 0.391
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.9497584541062802,
-  "best_model_checkpoint": "vit-msn-small-corect_deepcleaned_dataset_lateral_flow_ivalidation/checkpoint-22",
+  "best_metric": 0.9194139194139194,
+  "best_model_checkpoint": "vit-msn-small-corect_deepcleaned_dataset_lateral_flow_ivalidation/checkpoint-39",
   "epoch": 18.46153846153846,
   "eval_steps": 500,
   "global_step": 60,
@@ -10,225 +10,225 @@
   "log_history": [
     {
       "epoch": 0.9230769230769231,
-      "eval_accuracy": 0.8927536231884058,
-      "eval_loss": 0.3933483958244324,
-      "eval_runtime": 4.3694,
-      "eval_samples_per_second": 236.874,
-      "eval_steps_per_second": 3.891,
+      "eval_accuracy": 0.7216117216117216,
+      "eval_loss": 0.6175123453140259,
+      "eval_runtime": 1.1447,
+      "eval_samples_per_second": 238.5,
+      "eval_steps_per_second": 4.368,
       "step": 3
     },
     {
       "epoch": 1.8461538461538463,
-      "eval_accuracy": 0.9188405797101449,
-      "eval_loss": 0.4166622459888458,
-      "eval_runtime": 4.2818,
-      "eval_samples_per_second": 241.72,
-      "eval_steps_per_second": 3.97,
+      "eval_accuracy": 0.8351648351648352,
+      "eval_loss": 0.4141041934490204,
+      "eval_runtime": 1.1326,
+      "eval_samples_per_second": 241.031,
+      "eval_steps_per_second": 4.414,
       "step": 6
     },
     {
       "epoch": 2.769230769230769,
-      "eval_accuracy": 0.4966183574879227,
-      "eval_loss": 0.933142364025116,
-      "eval_runtime": 4.3066,
-      "eval_samples_per_second": 240.328,
-      "eval_steps_per_second": 3.947,
+      "eval_accuracy": 0.5787545787545788,
+      "eval_loss": 0.7407873868942261,
+      "eval_runtime": 1.1747,
+      "eval_samples_per_second": 232.409,
+      "eval_steps_per_second": 4.257,
       "step": 9
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 62.32957458496094,
+      "grad_norm": 70.46142578125,
       "learning_rate": 4.62962962962963e-05,
-      "loss": 0.615,
+      "loss": 0.5817,
       "step": 10
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9178743961352657,
-      "eval_loss": 0.3084805905818939,
-      "eval_runtime": 4.3796,
-      "eval_samples_per_second": 236.322,
-      "eval_steps_per_second": 3.882,
+      "eval_accuracy": 0.9157509157509157,
+      "eval_loss": 0.27566295862197876,
+      "eval_runtime": 1.1638,
+      "eval_samples_per_second": 234.573,
+      "eval_steps_per_second": 4.296,
       "step": 13
     },
     {
       "epoch": 4.923076923076923,
-      "eval_accuracy": 0.9333333333333333,
-      "eval_loss": 0.22097891569137573,
-      "eval_runtime": 4.3446,
-      "eval_samples_per_second": 238.226,
-      "eval_steps_per_second": 3.913,
+      "eval_accuracy": 0.8791208791208791,
+      "eval_loss": 0.28472262620925903,
+      "eval_runtime": 1.1571,
+      "eval_samples_per_second": 235.942,
+      "eval_steps_per_second": 4.321,
       "step": 16
     },
     {
       "epoch": 5.846153846153846,
-      "eval_accuracy": 0.9391304347826087,
-      "eval_loss": 0.21061654388904572,
-      "eval_runtime": 4.4485,
-      "eval_samples_per_second": 232.66,
-      "eval_steps_per_second": 3.821,
+      "eval_accuracy": 0.9010989010989011,
+      "eval_loss": 0.2455928921699524,
+      "eval_runtime": 1.2196,
+      "eval_samples_per_second": 223.848,
+      "eval_steps_per_second": 4.1,
       "step": 19
     },
     {
       "epoch": 6.153846153846154,
-      "grad_norm": 3.9477076530456543,
+      "grad_norm": 8.08311653137207,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.3763,
+      "loss": 0.3724,
       "step": 20
     },
     {
       "epoch": 6.769230769230769,
-      "eval_accuracy": 0.9497584541062802,
-      "eval_loss": 0.1871330887079239,
-      "eval_runtime": 4.3598,
-      "eval_samples_per_second": 237.396,
-      "eval_steps_per_second": 3.899,
+      "eval_accuracy": 0.9120879120879121,
+      "eval_loss": 0.25474727153778076,
+      "eval_runtime": 1.1827,
+      "eval_samples_per_second": 230.827,
+      "eval_steps_per_second": 4.228,
       "step": 22
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.9371980676328503,
-      "eval_loss": 0.20427261292934418,
-      "eval_runtime": 4.2827,
-      "eval_samples_per_second": 241.673,
-      "eval_steps_per_second": 3.97,
+      "eval_accuracy": 0.8827838827838828,
+      "eval_loss": 0.3007383942604065,
+      "eval_runtime": 1.2369,
+      "eval_samples_per_second": 220.714,
+      "eval_steps_per_second": 4.042,
       "step": 26
     },
     {
       "epoch": 8.923076923076923,
-      "eval_accuracy": 0.8888888888888888,
-      "eval_loss": 0.31205740571022034,
-      "eval_runtime": 4.4195,
-      "eval_samples_per_second": 234.192,
-      "eval_steps_per_second": 3.847,
+      "eval_accuracy": 0.9010989010989011,
+      "eval_loss": 0.3042505383491516,
+      "eval_runtime": 1.1883,
+      "eval_samples_per_second": 229.74,
+      "eval_steps_per_second": 4.208,
       "step": 29
     },
     {
       "epoch": 9.23076923076923,
-      "grad_norm": 4.070008754730225,
+      "grad_norm": 12.493155479431152,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 0.3511,
+      "loss": 0.3155,
       "step": 30
     },
     {
       "epoch": 9.846153846153847,
-      "eval_accuracy": 0.9314009661835749,
-      "eval_loss": 0.20154337584972382,
-      "eval_runtime": 4.2969,
-      "eval_samples_per_second": 240.871,
-      "eval_steps_per_second": 3.956,
+      "eval_accuracy": 0.9047619047619048,
+      "eval_loss": 0.2603397071361542,
+      "eval_runtime": 1.1963,
+      "eval_samples_per_second": 228.207,
+      "eval_steps_per_second": 4.18,
       "step": 32
     },
     {
       "epoch": 10.76923076923077,
-      "eval_accuracy": 0.8376811594202899,
-      "eval_loss": 0.44852134585380554,
-      "eval_runtime": 4.3709,
-      "eval_samples_per_second": 236.795,
-      "eval_steps_per_second": 3.889,
+      "eval_accuracy": 0.9157509157509157,
+      "eval_loss": 0.24810168147087097,
+      "eval_runtime": 1.1361,
+      "eval_samples_per_second": 240.289,
+      "eval_steps_per_second": 4.401,
       "step": 35
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.9285024154589372,
-      "eval_loss": 0.24452261626720428,
-      "eval_runtime": 4.3262,
-      "eval_samples_per_second": 239.241,
-      "eval_steps_per_second": 3.93,
+      "eval_accuracy": 0.9194139194139194,
+      "eval_loss": 0.22294031083583832,
+      "eval_runtime": 1.1914,
+      "eval_samples_per_second": 229.149,
+      "eval_steps_per_second": 4.197,
       "step": 39
     },
     {
       "epoch": 12.307692307692308,
-      "grad_norm": 7.704187393188477,
+      "grad_norm": 8.086904525756836,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.2962,
+      "loss": 0.2844,
       "step": 40
     },
     {
       "epoch": 12.923076923076923,
-      "eval_accuracy": 0.9053140096618357,
-      "eval_loss": 0.30447185039520264,
-      "eval_runtime": 4.2886,
-      "eval_samples_per_second": 241.336,
-      "eval_steps_per_second": 3.964,
+      "eval_accuracy": 0.8791208791208791,
+      "eval_loss": 0.3036467432975769,
+      "eval_runtime": 1.1286,
+      "eval_samples_per_second": 241.895,
+      "eval_steps_per_second": 4.43,
       "step": 42
     },
     {
       "epoch": 13.846153846153847,
-      "eval_accuracy": 0.8714975845410629,
-      "eval_loss": 0.3914734721183777,
-      "eval_runtime": 4.4006,
-      "eval_samples_per_second": 235.195,
-      "eval_steps_per_second": 3.863,
+      "eval_accuracy": 0.9084249084249084,
+      "eval_loss": 0.2578851580619812,
+      "eval_runtime": 1.1539,
+      "eval_samples_per_second": 236.595,
+      "eval_steps_per_second": 4.333,
       "step": 45
     },
     {
       "epoch": 14.76923076923077,
-      "eval_accuracy": 0.893719806763285,
-      "eval_loss": 0.31650951504707336,
-      "eval_runtime": 4.327,
-      "eval_samples_per_second": 239.195,
-      "eval_steps_per_second": 3.929,
+      "eval_accuracy": 0.9157509157509157,
+      "eval_loss": 0.2433992624282837,
+      "eval_runtime": 1.2067,
+      "eval_samples_per_second": 226.24,
+      "eval_steps_per_second": 4.144,
       "step": 48
     },
     {
       "epoch": 15.384615384615385,
-      "grad_norm": 5.901269912719727,
+      "grad_norm": 3.2125346660614014,
       "learning_rate": 9.259259259259259e-06,
-      "loss": 0.2553,
+      "loss": 0.2517,
       "step": 50
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.9082125603864735,
-      "eval_loss": 0.28225788474082947,
-      "eval_runtime": 4.4633,
-      "eval_samples_per_second": 231.89,
-      "eval_steps_per_second": 3.809,
+      "eval_accuracy": 0.9047619047619048,
+      "eval_loss": 0.27184122800827026,
+      "eval_runtime": 1.1306,
+      "eval_samples_per_second": 241.471,
+      "eval_steps_per_second": 4.423,
       "step": 52
     },
     {
       "epoch": 16.923076923076923,
-      "eval_accuracy": 0.8869565217391304,
-      "eval_loss": 0.3504292070865631,
-      "eval_runtime": 4.37,
-      "eval_samples_per_second": 236.841,
-      "eval_steps_per_second": 3.89,
+      "eval_accuracy": 0.9120879120879121,
+      "eval_loss": 0.25126636028289795,
+      "eval_runtime": 1.1542,
+      "eval_samples_per_second": 236.531,
+      "eval_steps_per_second": 4.332,
       "step": 55
     },
     {
       "epoch": 17.846153846153847,
-      "eval_accuracy": 0.8869565217391304,
-      "eval_loss": 0.36792051792144775,
-      "eval_runtime": 4.4082,
-      "eval_samples_per_second": 234.792,
-      "eval_steps_per_second": 3.856,
+      "eval_accuracy": 0.9120879120879121,
+      "eval_loss": 0.25034624338150024,
+      "eval_runtime": 1.2382,
+      "eval_samples_per_second": 220.484,
+      "eval_steps_per_second": 4.038,
       "step": 58
     },
     {
       "epoch": 18.46153846153846,
-      "grad_norm": 7.85857629776001,
+      "grad_norm": 4.7955803871154785,
       "learning_rate": 0.0,
-      "loss": 0.2601,
+      "loss": 0.2468,
       "step": 60
     },
     {
       "epoch": 18.46153846153846,
-      "eval_accuracy": 0.8879227053140096,
-      "eval_loss": 0.3520191013813019,
-      "eval_runtime": 4.3424,
-      "eval_samples_per_second": 238.347,
-      "eval_steps_per_second": 3.915,
+      "eval_accuracy": 0.9120879120879121,
+      "eval_loss": 0.24911320209503174,
+      "eval_runtime": 1.1404,
+      "eval_samples_per_second": 239.4,
+      "eval_steps_per_second": 4.385,
       "step": 60
     },
     {
       "epoch": 18.46153846153846,
       "step": 60,
       "total_flos": 2.9458314315627725e+17,
-      "train_loss": 0.35900469223658243,
-      "train_runtime": 212.0852,
-      "train_samples_per_second": 76.856,
-      "train_steps_per_second": 0.283
+      "train_loss": 0.342086935043335,
+      "train_runtime": 153.4118,
+      "train_samples_per_second": 106.25,
+      "train_steps_per_second": 0.391
     }
   ],
   "logging_steps": 10,