End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/beit-base-patch16-224-pt22k-ft22k](https://huggingface.co/microsoft/beit-base-patch16-224-pt22k-ft22k) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2845
-- Accuracy: 0.9058
+- Loss: 0.2796
+- Accuracy: 0.9082
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.99,
-    "eval_accuracy": 0.9192,
-    "eval_loss": 0.23591181635856628,
-    "eval_runtime": 78.5827,
-    "eval_samples_per_second": 63.627,
-    "eval_steps_per_second": 1.998
+    "epoch": 1.0,
+    "eval_accuracy": 0.9082,
+    "eval_loss": 0.2795782685279846,
+    "eval_runtime": 78.9983,
+    "eval_samples_per_second": 63.293,
+    "eval_steps_per_second": 1.987
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.99,
-    "eval_accuracy": 0.9192,
-    "eval_loss": 0.23591181635856628,
-    "eval_runtime": 78.5827,
-    "eval_samples_per_second": 63.627,
-    "eval_steps_per_second": 1.998
+    "epoch": 1.0,
+    "eval_accuracy": 0.9082,
+    "eval_loss": 0.2795782685279846,
+    "eval_runtime": 78.9983,
+    "eval_samples_per_second": 63.293,
+    "eval_steps_per_second": 1.987
 }

runs/Apr24_11-55-23_MacBook-Pro.local/events.out.tfevents.1713933335.MacBook-Pro.local.24885.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a0eebd7e731627a42f4dbd395cb5165d97d9aed588729b210142189bbe10031
+size 411

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9064,
+  "best_metric": 0.9058,
   "best_model_checkpoint": "beit-base-patch16-224-pt22k-ft22k-finetuning/checkpoint-351",
   "epoch": 0.997867803837953,
   "eval_steps": 500,
@@ -10,265 +10,265 @@
   "log_history": [
     {
       "epoch": 0.03,
-      "grad_norm": 12.392043113708496,
+      "grad_norm": 11.096900939941406,
       "learning_rate": 1.388888888888889e-05,
-      "loss": 2.38,
+      "loss": 2.3539,
       "step": 10
     },
     {
       "epoch": 0.06,
-      "grad_norm": 9.366357803344727,
+      "grad_norm": 8.86681079864502,
       "learning_rate": 2.777777777777778e-05,
-      "loss": 2.0468,
+      "loss": 2.0154,
       "step": 20
     },
     {
       "epoch": 0.09,
-      "grad_norm": 10.620584487915039,
+      "grad_norm": 9.853775978088379,
       "learning_rate": 4.166666666666667e-05,
-      "loss": 1.5116,
+      "loss": 1.4709,
       "step": 30
     },
     {
       "epoch": 0.11,
-      "grad_norm": 9.78640079498291,
+      "grad_norm": 10.110260963439941,
       "learning_rate": 4.936507936507937e-05,
-      "loss": 1.0056,
+      "loss": 1.0243,
       "step": 40
     },
     {
       "epoch": 0.14,
-      "grad_norm": 7.943943977355957,
+      "grad_norm": 10.437296867370605,
       "learning_rate": 4.7777777777777784e-05,
-      "loss": 0.7888,
+      "loss": 0.7642,
       "step": 50
     },
     {
       "epoch": 0.17,
-      "grad_norm": 9.149613380432129,
+      "grad_norm": 7.33429479598999,
       "learning_rate": 4.6190476190476194e-05,
-      "loss": 0.6486,
+      "loss": 0.6573,
       "step": 60
     },
     {
       "epoch": 0.2,
-      "grad_norm": 7.764841079711914,
+      "grad_norm": 7.737111568450928,
       "learning_rate": 4.460317460317461e-05,
-      "loss": 0.6232,
+      "loss": 0.5998,
       "step": 70
     },
     {
       "epoch": 0.23,
-      "grad_norm": 8.434979438781738,
+      "grad_norm": 7.117513179779053,
       "learning_rate": 4.301587301587302e-05,
-      "loss": 0.5555,
+      "loss": 0.5478,
       "step": 80
     },
     {
       "epoch": 0.26,
-      "grad_norm": 8.733748435974121,
+      "grad_norm": 8.698220252990723,
       "learning_rate": 4.1428571428571437e-05,
-      "loss": 0.4872,
+      "loss": 0.5195,
       "step": 90
     },
     {
       "epoch": 0.28,
-      "grad_norm": 7.663728713989258,
+      "grad_norm": 6.8238677978515625,
       "learning_rate": 3.984126984126984e-05,
-      "loss": 0.5385,
+      "loss": 0.5212,
       "step": 100
     },
     {
       "epoch": 0.31,
-      "grad_norm": 6.9408698081970215,
+      "grad_norm": 6.604701042175293,
       "learning_rate": 3.8253968253968256e-05,
-      "loss": 0.4591,
+      "loss": 0.4546,
       "step": 110
     },
     {
       "epoch": 0.34,
-      "grad_norm": 6.367518424987793,
+      "grad_norm": 8.972763061523438,
       "learning_rate": 3.6666666666666666e-05,
-      "loss": 0.4449,
+      "loss": 0.4429,
       "step": 120
     },
     {
       "epoch": 0.37,
-      "grad_norm": 6.589195728302002,
+      "grad_norm": 6.5601959228515625,
       "learning_rate": 3.5079365079365075e-05,
-      "loss": 0.4365,
+      "loss": 0.4326,
       "step": 130
     },
     {
       "epoch": 0.4,
-      "grad_norm": 8.5034818649292,
+      "grad_norm": 7.662568092346191,
       "learning_rate": 3.349206349206349e-05,
-      "loss": 0.4538,
+      "loss": 0.4859,
       "step": 140
     },
     {
       "epoch": 0.43,
-      "grad_norm": 6.432806491851807,
+      "grad_norm": 8.094512939453125,
       "learning_rate": 3.19047619047619e-05,
-      "loss": 0.4182,
+      "loss": 0.4159,
       "step": 150
     },
     {
       "epoch": 0.45,
-      "grad_norm": 5.926751613616943,
+      "grad_norm": 6.0086846351623535,
       "learning_rate": 3.0317460317460318e-05,
-      "loss": 0.3683,
+      "loss": 0.4216,
       "step": 160
     },
     {
       "epoch": 0.48,
-      "grad_norm": 7.516120433807373,
+      "grad_norm": 7.797686576843262,
       "learning_rate": 2.8730158730158728e-05,
-      "loss": 0.4315,
+      "loss": 0.434,
       "step": 170
     },
     {
       "epoch": 0.51,
-      "grad_norm": 7.042306423187256,
+      "grad_norm": 7.188005447387695,
       "learning_rate": 2.714285714285714e-05,
-      "loss": 0.3883,
+      "loss": 0.3736,
       "step": 180
     },
     {
       "epoch": 0.54,
-      "grad_norm": 6.7353620529174805,
+      "grad_norm": 5.677178382873535,
       "learning_rate": 2.5555555555555554e-05,
-      "loss": 0.4158,
+      "loss": 0.3726,
       "step": 190
     },
     {
       "epoch": 0.57,
-      "grad_norm": 6.9553914070129395,
+      "grad_norm": 8.169827461242676,
       "learning_rate": 2.396825396825397e-05,
-      "loss": 0.3893,
+      "loss": 0.3675,
       "step": 200
     },
     {
       "epoch": 0.6,
-      "grad_norm": 6.948991298675537,
+      "grad_norm": 6.677453994750977,
       "learning_rate": 2.2380952380952384e-05,
-      "loss": 0.4288,
+      "loss": 0.3794,
       "step": 210
     },
     {
       "epoch": 0.63,
-      "grad_norm": 6.90835428237915,
+      "grad_norm": 6.1892876625061035,
       "learning_rate": 2.0793650793650797e-05,
-      "loss": 0.368,
+      "loss": 0.4131,
       "step": 220
     },
     {
       "epoch": 0.65,
-      "grad_norm": 7.47681999206543,
+      "grad_norm": 7.120225429534912,
       "learning_rate": 1.920634920634921e-05,
-      "loss": 0.4212,
+      "loss": 0.3614,
       "step": 230
     },
     {
       "epoch": 0.68,
-      "grad_norm": 5.4326887130737305,
+      "grad_norm": 7.643838405609131,
       "learning_rate": 1.761904761904762e-05,
-      "loss": 0.4101,
+      "loss": 0.3806,
       "step": 240
     },
     {
       "epoch": 0.71,
-      "grad_norm": 5.068437576293945,
+      "grad_norm": 5.949714660644531,
       "learning_rate": 1.6031746031746033e-05,
-      "loss": 0.3544,
+      "loss": 0.33,
       "step": 250
     },
     {
       "epoch": 0.74,
-      "grad_norm": 5.686431884765625,
+      "grad_norm": 7.846484661102295,
       "learning_rate": 1.4444444444444444e-05,
-      "loss": 0.3252,
+      "loss": 0.3952,
       "step": 260
     },
     {
       "epoch": 0.77,
-      "grad_norm": 5.882190227508545,
+      "grad_norm": 8.437196731567383,
       "learning_rate": 1.2857142857142857e-05,
-      "loss": 0.3475,
+      "loss": 0.3512,
       "step": 270
     },
     {
       "epoch": 0.8,
-      "grad_norm": 6.3953447341918945,
+      "grad_norm": 6.006718158721924,
       "learning_rate": 1.126984126984127e-05,
-      "loss": 0.3164,
+      "loss": 0.3003,
       "step": 280
     },
     {
       "epoch": 0.82,
-      "grad_norm": 7.278485298156738,
+      "grad_norm": 8.551034927368164,
       "learning_rate": 9.682539682539683e-06,
-      "loss": 0.328,
+      "loss": 0.3413,
       "step": 290
     },
     {
       "epoch": 0.85,
-      "grad_norm": 6.7763166427612305,
+      "grad_norm": 7.815725326538086,
       "learning_rate": 8.095238095238097e-06,
-      "loss": 0.3463,
+      "loss": 0.3174,
       "step": 300
     },
     {
       "epoch": 0.88,
-      "grad_norm": 7.21019172668457,
+      "grad_norm": 7.373149871826172,
       "learning_rate": 6.507936507936509e-06,
-      "loss": 0.3762,
+      "loss": 0.337,
       "step": 310
     },
     {
       "epoch": 0.91,
-      "grad_norm": 8.0834379196167,
+      "grad_norm": 6.722525596618652,
       "learning_rate": 4.920634920634921e-06,
-      "loss": 0.3538,
+      "loss": 0.3622,
       "step": 320
     },
     {
       "epoch": 0.94,
-      "grad_norm": 5.969318389892578,
+      "grad_norm": 8.15896224975586,
       "learning_rate": 3.3333333333333333e-06,
-      "loss": 0.3159,
+      "loss": 0.3043,
       "step": 330
     },
     {
       "epoch": 0.97,
-      "grad_norm": 7.134821891784668,
+      "grad_norm": 6.256545066833496,
       "learning_rate": 1.7460317460317462e-06,
-      "loss": 0.3528,
+      "loss": 0.3147,
       "step": 340
     },
     {
       "epoch": 1.0,
-      "grad_norm": 6.126035213470459,
+      "grad_norm": 6.901019096374512,
       "learning_rate": 1.5873015873015874e-07,
-      "loss": 0.2928,
+      "loss": 0.3075,
       "step": 350
     },
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.9064,
-      "eval_loss": 0.27925798296928406,
-      "eval_runtime": 77.4382,
-      "eval_samples_per_second": 64.568,
-      "eval_steps_per_second": 2.027,
+      "eval_accuracy": 0.9058,
+      "eval_loss": 0.2844555079936981,
+      "eval_runtime": 80.5038,
+      "eval_samples_per_second": 62.109,
+      "eval_steps_per_second": 1.95,
       "step": 351
     },
     {
       "epoch": 1.0,
       "step": 351,
       "total_flos": 3.4803216831306793e+18,
-      "train_loss": 0.5742807560666674,
-      "train_runtime": 2256.218,
-      "train_samples_per_second": 19.945,
+      "train_loss": 0.5666828784253183,
+      "train_runtime": 2255.5076,
+      "train_samples_per_second": 19.951,
       "train_steps_per_second": 0.156
     }
   ],