End of training

README.md

@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2087
+- Loss: 1.6325
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 4.29345703125,
-    "eval_runtime": 1.1017,
-    "eval_samples_per_second": 41.752,
-    "eval_steps_per_second": 2.723,
-    "train_loss": 2.6118680079778036,
-    "train_runtime": 200.6839,
-    "train_samples_per_second": 42.455,
-    "train_steps_per_second": 0.598
+    "eval_loss": 1.6324856281280518,
+    "eval_runtime": 0.9999,
+    "eval_samples_per_second": 46.003,
+    "eval_steps_per_second": 3.0,
+    "train_loss": 3.0923615217208864,
+    "train_runtime": 200.015,
+    "train_samples_per_second": 42.597,
+    "train_steps_per_second": 0.6
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 4.29345703125,
-    "eval_runtime": 1.1017,
-    "eval_samples_per_second": 41.752,
-    "eval_steps_per_second": 2.723
+    "eval_loss": 1.6324856281280518,
+    "eval_runtime": 0.9999,
+    "eval_samples_per_second": 46.003,
+    "eval_steps_per_second": 3.0
 }

runs/Dec02_14-25-25_DESKTOP-SKBE9FB/events.out.tfevents.1733171402.DESKTOP-SKBE9FB.14404.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75f98cf67524644b96e1bc678367ce2e74385ac06db7ad2e997a2c00643b82b0
+size 405

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 34.29,
-    "train_loss": 2.6118680079778036,
-    "train_runtime": 200.6839,
-    "train_samples_per_second": 42.455,
-    "train_steps_per_second": 0.598
+    "train_loss": 3.0923615217208864,
+    "train_runtime": 200.015,
+    "train_samples_per_second": 42.597,
+    "train_steps_per_second": 0.6
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.45652173913043476,
-  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\checkpoint-21",
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\checkpoint-63",
   "epoch": 34.285714285714285,
   "eval_steps": 500,
   "global_step": 120,
@@ -11,398 +11,398 @@
     {
       "epoch": 0.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 4.433082580566406,
-      "eval_runtime": 0.8144,
-      "eval_samples_per_second": 56.485,
-      "eval_steps_per_second": 3.684,
+      "eval_loss": 7.935995578765869,
+      "eval_runtime": 0.8407,
+      "eval_samples_per_second": 54.717,
+      "eval_steps_per_second": 3.569,
       "step": 3
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 39.162105560302734,
-      "eval_runtime": 0.7581,
-      "eval_samples_per_second": 60.681,
-      "eval_steps_per_second": 3.957,
+      "eval_loss": 7.836058139801025,
+      "eval_runtime": 0.7656,
+      "eval_samples_per_second": 60.085,
+      "eval_steps_per_second": 3.919,
       "step": 7
     },
     {
       "epoch": 2.86,
-      "learning_rate": 0.075,
-      "loss": 9.3798,
+      "learning_rate": 3.859649122807018e-05,
+      "loss": 7.9038,
       "step": 10
     },
     {
       "epoch": 2.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 7.2709150314331055,
-      "eval_runtime": 0.7621,
-      "eval_samples_per_second": 60.363,
-      "eval_steps_per_second": 3.937,
+      "eval_loss": 7.584959506988525,
+      "eval_runtime": 0.899,
+      "eval_samples_per_second": 51.168,
+      "eval_steps_per_second": 3.337,
       "step": 10
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 4.332581996917725,
-      "eval_runtime": 0.7692,
-      "eval_samples_per_second": 59.804,
-      "eval_steps_per_second": 3.9,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.062610626220703,
+      "eval_runtime": 0.7578,
+      "eval_samples_per_second": 60.698,
+      "eval_steps_per_second": 3.959,
       "step": 14
     },
     {
       "epoch": 4.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 6.965109825134277,
-      "eval_runtime": 0.7683,
-      "eval_samples_per_second": 59.876,
-      "eval_steps_per_second": 3.905,
+      "eval_loss": 6.55360221862793,
+      "eval_runtime": 0.7578,
+      "eval_samples_per_second": 60.703,
+      "eval_steps_per_second": 3.959,
       "step": 17
     },
     {
       "epoch": 5.71,
-      "learning_rate": 0.08333333333333333,
-      "loss": 6.904,
+      "learning_rate": 3.508771929824562e-05,
+      "loss": 6.9398,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 4.29345703125,
-      "eval_runtime": 0.8512,
-      "eval_samples_per_second": 54.04,
-      "eval_steps_per_second": 3.524,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 5.9112935066223145,
+      "eval_runtime": 0.7546,
+      "eval_samples_per_second": 60.958,
+      "eval_steps_per_second": 3.976,
       "step": 21
     },
     {
       "epoch": 6.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 3.313812255859375,
-      "eval_runtime": 0.7552,
-      "eval_samples_per_second": 60.911,
-      "eval_steps_per_second": 3.972,
+      "eval_loss": 5.422503471374512,
+      "eval_runtime": 0.7483,
+      "eval_samples_per_second": 61.476,
+      "eval_steps_per_second": 4.009,
       "step": 24
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 3.5143587589263916,
-      "eval_runtime": 0.7666,
-      "eval_samples_per_second": 60.004,
-      "eval_steps_per_second": 3.913,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 4.840947151184082,
+      "eval_runtime": 0.7567,
+      "eval_samples_per_second": 60.791,
+      "eval_steps_per_second": 3.965,
       "step": 28
     },
     {
       "epoch": 8.57,
-      "learning_rate": 0.075,
-      "loss": 2.7643,
+      "learning_rate": 3.157894736842106e-05,
+      "loss": 5.3358,
       "step": 30
     },
     {
       "epoch": 8.86,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 4.460618495941162,
-      "eval_runtime": 0.7671,
-      "eval_samples_per_second": 59.964,
-      "eval_steps_per_second": 3.911,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 4.419978141784668,
+      "eval_runtime": 0.7252,
+      "eval_samples_per_second": 63.432,
+      "eval_steps_per_second": 4.137,
       "step": 31
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.648934245109558,
-      "eval_runtime": 0.7937,
-      "eval_samples_per_second": 57.955,
-      "eval_steps_per_second": 3.78,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 3.8920633792877197,
+      "eval_runtime": 0.8521,
+      "eval_samples_per_second": 53.987,
+      "eval_steps_per_second": 3.521,
       "step": 35
     },
     {
       "epoch": 10.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.3025492429733276,
-      "eval_runtime": 0.7632,
-      "eval_samples_per_second": 60.275,
-      "eval_steps_per_second": 3.931,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 3.5183205604553223,
+      "eval_runtime": 0.7662,
+      "eval_samples_per_second": 60.038,
+      "eval_steps_per_second": 3.916,
       "step": 38
     },
     {
       "epoch": 11.43,
-      "learning_rate": 0.06666666666666667,
-      "loss": 2.4694,
+      "learning_rate": 2.8070175438596492e-05,
+      "loss": 4.1182,
       "step": 40
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.4723820686340332,
-      "eval_runtime": 0.7612,
-      "eval_samples_per_second": 60.433,
-      "eval_steps_per_second": 3.941,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 3.059624671936035,
+      "eval_runtime": 0.8767,
+      "eval_samples_per_second": 52.468,
+      "eval_steps_per_second": 3.422,
       "step": 42
     },
     {
       "epoch": 12.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.359671711921692,
-      "eval_runtime": 0.7647,
-      "eval_samples_per_second": 60.156,
-      "eval_steps_per_second": 3.923,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 2.752983808517456,
+      "eval_runtime": 0.7467,
+      "eval_samples_per_second": 61.606,
+      "eval_steps_per_second": 4.018,
       "step": 45
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2714816331863403,
-      "eval_runtime": 0.7564,
-      "eval_samples_per_second": 60.811,
-      "eval_steps_per_second": 3.966,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 2.397618055343628,
+      "eval_runtime": 0.7077,
+      "eval_samples_per_second": 65.002,
+      "eval_steps_per_second": 4.239,
       "step": 49
     },
     {
       "epoch": 14.29,
-      "learning_rate": 0.05833333333333333,
-      "loss": 1.3638,
+      "learning_rate": 2.4561403508771932e-05,
+      "loss": 2.8935,
       "step": 50
     },
     {
       "epoch": 14.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2384672164916992,
-      "eval_runtime": 0.7792,
-      "eval_samples_per_second": 59.031,
-      "eval_steps_per_second": 3.85,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 2.1733179092407227,
+      "eval_runtime": 0.7167,
+      "eval_samples_per_second": 64.184,
+      "eval_steps_per_second": 4.186,
       "step": 52
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.217814564704895,
-      "eval_runtime": 0.796,
-      "eval_samples_per_second": 57.79,
-      "eval_steps_per_second": 3.769,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1.9295960664749146,
+      "eval_runtime": 0.7447,
+      "eval_samples_per_second": 61.772,
+      "eval_steps_per_second": 4.029,
       "step": 56
     },
     {
       "epoch": 16.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.228389859199524,
-      "eval_runtime": 0.7751,
-      "eval_samples_per_second": 59.345,
-      "eval_steps_per_second": 3.87,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1.7845752239227295,
+      "eval_runtime": 0.7372,
+      "eval_samples_per_second": 62.401,
+      "eval_steps_per_second": 4.07,
       "step": 59
     },
     {
       "epoch": 17.14,
-      "learning_rate": 0.05,
-      "loss": 1.2334,
+      "learning_rate": 2.105263157894737e-05,
+      "loss": 2.0579,
       "step": 60
     },
     {
       "epoch": 18.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2211333513259888,
-      "eval_runtime": 0.7381,
-      "eval_samples_per_second": 62.321,
-      "eval_steps_per_second": 4.064,
+      "eval_loss": 1.6324856281280518,
+      "eval_runtime": 0.7287,
+      "eval_samples_per_second": 63.129,
+      "eval_steps_per_second": 4.117,
       "step": 63
     },
     {
       "epoch": 18.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2433996200561523,
-      "eval_runtime": 0.7676,
-      "eval_samples_per_second": 59.923,
-      "eval_steps_per_second": 3.908,
+      "eval_loss": 1.5432193279266357,
+      "eval_runtime": 0.7462,
+      "eval_samples_per_second": 61.648,
+      "eval_steps_per_second": 4.021,
       "step": 66
     },
     {
       "epoch": 20.0,
-      "learning_rate": 0.041666666666666664,
-      "loss": 1.2262,
+      "learning_rate": 1.754385964912281e-05,
+      "loss": 1.6272,
       "step": 70
     },
     {
       "epoch": 20.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2208575010299683,
-      "eval_runtime": 0.7578,
-      "eval_samples_per_second": 60.699,
-      "eval_steps_per_second": 3.959,
+      "eval_loss": 1.4466660022735596,
+      "eval_runtime": 0.7257,
+      "eval_samples_per_second": 63.39,
+      "eval_steps_per_second": 4.134,
       "step": 70
     },
     {
       "epoch": 20.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2129850387573242,
-      "eval_runtime": 0.7544,
-      "eval_samples_per_second": 60.974,
-      "eval_steps_per_second": 3.977,
+      "eval_loss": 1.386461615562439,
+      "eval_runtime": 0.7507,
+      "eval_samples_per_second": 61.278,
+      "eval_steps_per_second": 3.996,
       "step": 73
     },
     {
       "epoch": 22.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2383720874786377,
-      "eval_runtime": 0.853,
-      "eval_samples_per_second": 53.926,
-      "eval_steps_per_second": 3.517,
+      "eval_loss": 1.32200026512146,
+      "eval_runtime": 0.8387,
+      "eval_samples_per_second": 54.846,
+      "eval_steps_per_second": 3.577,
       "step": 77
     },
     {
       "epoch": 22.86,
-      "learning_rate": 0.03333333333333333,
-      "loss": 1.2038,
+      "learning_rate": 1.4035087719298246e-05,
+      "loss": 1.3699,
       "step": 80
     },
     {
       "epoch": 22.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2312527894973755,
-      "eval_runtime": 0.8452,
-      "eval_samples_per_second": 54.425,
-      "eval_steps_per_second": 3.549,
+      "eval_loss": 1.2847795486450195,
+      "eval_runtime": 1.0743,
+      "eval_samples_per_second": 42.82,
+      "eval_steps_per_second": 2.793,
       "step": 80
     },
     {
       "epoch": 24.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2086050510406494,
-      "eval_runtime": 0.8429,
-      "eval_samples_per_second": 54.571,
-      "eval_steps_per_second": 3.559,
+      "eval_loss": 1.2489309310913086,
+      "eval_runtime": 0.8362,
+      "eval_samples_per_second": 55.01,
+      "eval_steps_per_second": 3.588,
       "step": 84
     },
     {
       "epoch": 24.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2106943130493164,
-      "eval_runtime": 0.7507,
-      "eval_samples_per_second": 61.274,
-      "eval_steps_per_second": 3.996,
+      "eval_loss": 1.2313342094421387,
+      "eval_runtime": 0.7937,
+      "eval_samples_per_second": 57.955,
+      "eval_steps_per_second": 3.78,
       "step": 87
     },
     {
       "epoch": 25.71,
-      "learning_rate": 0.025,
-      "loss": 1.1975,
+      "learning_rate": 1.0526315789473684e-05,
+      "loss": 1.2469,
       "step": 90
     },
     {
       "epoch": 26.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2075337171554565,
-      "eval_runtime": 0.7507,
-      "eval_samples_per_second": 61.28,
-      "eval_steps_per_second": 3.997,
+      "eval_loss": 1.2175383567810059,
+      "eval_runtime": 0.7492,
+      "eval_samples_per_second": 61.401,
+      "eval_steps_per_second": 4.004,
       "step": 91
     },
     {
       "epoch": 26.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2106044292449951,
-      "eval_runtime": 0.7642,
-      "eval_samples_per_second": 60.195,
-      "eval_steps_per_second": 3.926,
+      "eval_loss": 1.2125990390777588,
+      "eval_runtime": 0.7412,
+      "eval_samples_per_second": 62.063,
+      "eval_steps_per_second": 4.048,
       "step": 94
     },
     {
       "epoch": 28.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2139972448349,
-      "eval_runtime": 0.8793,
-      "eval_samples_per_second": 52.316,
-      "eval_steps_per_second": 3.412,
+      "eval_loss": 1.2115219831466675,
+      "eval_runtime": 0.8532,
+      "eval_samples_per_second": 53.915,
+      "eval_steps_per_second": 3.516,
       "step": 98
     },
     {
       "epoch": 28.57,
-      "learning_rate": 0.016666666666666666,
-      "loss": 1.2005,
+      "learning_rate": 7.017543859649123e-06,
+      "loss": 1.2128,
       "step": 100
     },
     {
       "epoch": 28.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2094104290008545,
-      "eval_runtime": 0.7512,
-      "eval_samples_per_second": 61.234,
-      "eval_steps_per_second": 3.994,
+      "eval_loss": 1.2108038663864136,
+      "eval_runtime": 0.7442,
+      "eval_samples_per_second": 61.813,
+      "eval_steps_per_second": 4.031,
       "step": 101
     },
     {
       "epoch": 30.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2090147733688354,
-      "eval_runtime": 0.7768,
-      "eval_samples_per_second": 59.215,
-      "eval_steps_per_second": 3.862,
+      "eval_loss": 1.2107146978378296,
+      "eval_runtime": 0.7702,
+      "eval_samples_per_second": 59.727,
+      "eval_steps_per_second": 3.895,
       "step": 105
     },
     {
       "epoch": 30.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2084453105926514,
-      "eval_runtime": 0.7843,
-      "eval_samples_per_second": 58.654,
-      "eval_steps_per_second": 3.825,
+      "eval_loss": 1.2101784944534302,
+      "eval_runtime": 0.8207,
+      "eval_samples_per_second": 56.05,
+      "eval_steps_per_second": 3.655,
       "step": 108
     },
     {
       "epoch": 31.43,
-      "learning_rate": 0.008333333333333333,
-      "loss": 1.2012,
+      "learning_rate": 3.5087719298245615e-06,
+      "loss": 1.1889,
       "step": 110
     },
     {
       "epoch": 32.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2072227001190186,
-      "eval_runtime": 0.7702,
-      "eval_samples_per_second": 59.724,
-      "eval_steps_per_second": 3.895,
+      "eval_loss": 1.209415316581726,
+      "eval_runtime": 0.8748,
+      "eval_samples_per_second": 52.586,
+      "eval_steps_per_second": 3.43,
       "step": 112
     },
     {
       "epoch": 32.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2066519260406494,
-      "eval_runtime": 0.7797,
-      "eval_samples_per_second": 58.997,
-      "eval_steps_per_second": 3.848,
+      "eval_loss": 1.208990454673767,
+      "eval_runtime": 0.7574,
+      "eval_samples_per_second": 60.738,
+      "eval_steps_per_second": 3.961,
       "step": 115
     },
     {
       "epoch": 34.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2065129280090332,
-      "eval_runtime": 0.7602,
-      "eval_samples_per_second": 60.509,
-      "eval_steps_per_second": 3.946,
+      "eval_loss": 1.2087311744689941,
+      "eval_runtime": 0.7609,
+      "eval_samples_per_second": 60.453,
+      "eval_steps_per_second": 3.943,
       "step": 119
     },
     {
       "epoch": 34.29,
       "learning_rate": 0.0,
-      "loss": 1.1985,
+      "loss": 1.2137,
       "step": 120
     },
     {
       "epoch": 34.29,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2065057754516602,
-      "eval_runtime": 0.7543,
-      "eval_samples_per_second": 60.98,
-      "eval_steps_per_second": 3.977,
+      "eval_loss": 1.208715796470642,
+      "eval_runtime": 0.7628,
+      "eval_samples_per_second": 60.304,
+      "eval_steps_per_second": 3.933,
       "step": 120
     },
     {
       "epoch": 34.29,
       "step": 120,
       "total_flos": 2.3770905934823424e+17,
-      "train_loss": 2.6118680079778036,
-      "train_runtime": 200.6839,
-      "train_samples_per_second": 42.455,
-      "train_steps_per_second": 0.598
+      "train_loss": 3.0923615217208864,
+      "train_runtime": 200.015,
+      "train_samples_per_second": 42.597,
+      "train_steps_per_second": 0.6
     }
   ],
   "logging_steps": 10,