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: 1.2067
+- Loss: 2.8351
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "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,
+    "eval_loss": 2.8350560665130615,
+    "eval_runtime": 1.1289,
+    "eval_samples_per_second": 40.749,
+    "eval_steps_per_second": 2.658,
+    "train_loss": 2.340296713511149,
+    "train_runtime": 200.1533,
+    "train_samples_per_second": 42.567,
     "train_steps_per_second": 0.6
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 1.6324856281280518,
-    "eval_runtime": 0.9999,
-    "eval_samples_per_second": 46.003,
-    "eval_steps_per_second": 3.0
+    "eval_loss": 2.8350560665130615,
+    "eval_runtime": 1.1289,
+    "eval_samples_per_second": 40.749,
+    "eval_steps_per_second": 2.658
 }

runs/Dec02_14-35-31_DESKTOP-SKBE9FB/events.out.tfevents.1733172001.DESKTOP-SKBE9FB.18464.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5bbb61cb2398b07a9d84da8aac6e079ad51ec1baf0b0feccd7c646e7fc45397e
+size 405

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 34.29,
-    "train_loss": 3.0923615217208864,
-    "train_runtime": 200.015,
-    "train_samples_per_second": 42.597,
+    "train_loss": 2.340296713511149,
+    "train_runtime": 200.1533,
+    "train_samples_per_second": 42.567,
     "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-63",
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\checkpoint-7",
   "epoch": 34.285714285714285,
   "eval_steps": 500,
   "global_step": 120,
@@ -11,397 +11,397 @@
     {
       "epoch": 0.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 7.935995578765869,
-      "eval_runtime": 0.8407,
-      "eval_samples_per_second": 54.717,
-      "eval_steps_per_second": 3.569,
+      "eval_loss": 3.488030195236206,
+      "eval_runtime": 1.0513,
+      "eval_samples_per_second": 43.757,
+      "eval_steps_per_second": 2.854,
       "step": 3
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 7.836058139801025,
-      "eval_runtime": 0.7656,
-      "eval_samples_per_second": 60.085,
-      "eval_steps_per_second": 3.919,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.8350560665130615,
+      "eval_runtime": 0.7512,
+      "eval_samples_per_second": 61.235,
+      "eval_steps_per_second": 3.994,
       "step": 7
     },
     {
       "epoch": 2.86,
-      "learning_rate": 3.859649122807018e-05,
-      "loss": 7.9038,
+      "learning_rate": 0.04166666666666667,
+      "loss": 9.7554,
       "step": 10
     },
     {
       "epoch": 2.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 7.584959506988525,
-      "eval_runtime": 0.899,
-      "eval_samples_per_second": 51.168,
-      "eval_steps_per_second": 3.337,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 3.851775884628296,
+      "eval_runtime": 1.1058,
+      "eval_samples_per_second": 41.6,
+      "eval_steps_per_second": 2.713,
       "step": 10
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 7.062610626220703,
-      "eval_runtime": 0.7578,
-      "eval_samples_per_second": 60.698,
-      "eval_steps_per_second": 3.959,
+      "eval_loss": 2.7397520542144775,
+      "eval_runtime": 0.7536,
+      "eval_samples_per_second": 61.039,
+      "eval_steps_per_second": 3.981,
       "step": 14
     },
     {
       "epoch": 4.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 6.55360221862793,
-      "eval_runtime": 0.7578,
-      "eval_samples_per_second": 60.703,
-      "eval_steps_per_second": 3.959,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 4.487335681915283,
+      "eval_runtime": 0.7686,
+      "eval_samples_per_second": 59.851,
+      "eval_steps_per_second": 3.903,
       "step": 17
     },
     {
       "epoch": 5.71,
-      "learning_rate": 3.508771929824562e-05,
-      "loss": 6.9398,
+      "learning_rate": 0.0462962962962963,
+      "loss": 3.9513,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 5.9112935066223145,
-      "eval_runtime": 0.7546,
-      "eval_samples_per_second": 60.958,
-      "eval_steps_per_second": 3.976,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 3.855928421020508,
+      "eval_runtime": 0.7493,
+      "eval_samples_per_second": 61.394,
+      "eval_steps_per_second": 4.004,
       "step": 21
     },
     {
       "epoch": 6.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 5.422503471374512,
-      "eval_runtime": 0.7483,
-      "eval_samples_per_second": 61.476,
-      "eval_steps_per_second": 4.009,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.14131760597229,
+      "eval_runtime": 0.7545,
+      "eval_samples_per_second": 60.964,
+      "eval_steps_per_second": 3.976,
       "step": 24
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 4.840947151184082,
-      "eval_runtime": 0.7567,
-      "eval_samples_per_second": 60.791,
-      "eval_steps_per_second": 3.965,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.4669885635375977,
+      "eval_runtime": 0.7599,
+      "eval_samples_per_second": 60.535,
+      "eval_steps_per_second": 3.948,
       "step": 28
     },
     {
       "epoch": 8.57,
-      "learning_rate": 3.157894736842106e-05,
-      "loss": 5.3358,
+      "learning_rate": 0.04166666666666667,
+      "loss": 2.7557,
       "step": 30
     },
     {
       "epoch": 8.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 4.419978141784668,
-      "eval_runtime": 0.7252,
-      "eval_samples_per_second": 63.432,
-      "eval_steps_per_second": 4.137,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.352333426475525,
+      "eval_runtime": 0.7315,
+      "eval_samples_per_second": 62.884,
+      "eval_steps_per_second": 4.101,
       "step": 31
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 3.8920633792877197,
-      "eval_runtime": 0.8521,
-      "eval_samples_per_second": 53.987,
-      "eval_steps_per_second": 3.521,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.415340781211853,
+      "eval_runtime": 0.7434,
+      "eval_samples_per_second": 61.88,
+      "eval_steps_per_second": 4.036,
       "step": 35
     },
     {
       "epoch": 10.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 3.5183205604553223,
-      "eval_runtime": 0.7662,
-      "eval_samples_per_second": 60.038,
-      "eval_steps_per_second": 3.916,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.3797227144241333,
+      "eval_runtime": 0.7782,
+      "eval_samples_per_second": 59.107,
+      "eval_steps_per_second": 3.855,
       "step": 38
     },
     {
       "epoch": 11.43,
-      "learning_rate": 2.8070175438596492e-05,
-      "loss": 4.1182,
+      "learning_rate": 0.037037037037037035,
+      "loss": 1.6168,
       "step": 40
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 3.059624671936035,
-      "eval_runtime": 0.8767,
-      "eval_samples_per_second": 52.468,
-      "eval_steps_per_second": 3.422,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.4730011224746704,
+      "eval_runtime": 0.7535,
+      "eval_samples_per_second": 61.049,
+      "eval_steps_per_second": 3.981,
       "step": 42
     },
     {
       "epoch": 12.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 2.752983808517456,
-      "eval_runtime": 0.7467,
-      "eval_samples_per_second": 61.606,
-      "eval_steps_per_second": 4.018,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2619987726211548,
+      "eval_runtime": 0.734,
+      "eval_samples_per_second": 62.674,
+      "eval_steps_per_second": 4.087,
       "step": 45
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 2.397618055343628,
-      "eval_runtime": 0.7077,
-      "eval_samples_per_second": 65.002,
-      "eval_steps_per_second": 4.239,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2209850549697876,
+      "eval_runtime": 0.7662,
+      "eval_samples_per_second": 60.038,
+      "eval_steps_per_second": 3.915,
       "step": 49
     },
     {
       "epoch": 14.29,
-      "learning_rate": 2.4561403508771932e-05,
-      "loss": 2.8935,
+      "learning_rate": 0.032407407407407406,
+      "loss": 1.4447,
       "step": 50
     },
     {
       "epoch": 14.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 2.1733179092407227,
-      "eval_runtime": 0.7167,
-      "eval_samples_per_second": 64.184,
-      "eval_steps_per_second": 4.186,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2587504386901855,
+      "eval_runtime": 0.7387,
+      "eval_samples_per_second": 62.274,
+      "eval_steps_per_second": 4.061,
       "step": 52
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 1.9295960664749146,
-      "eval_runtime": 0.7447,
-      "eval_samples_per_second": 61.772,
-      "eval_steps_per_second": 4.029,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2458689212799072,
+      "eval_runtime": 0.7367,
+      "eval_samples_per_second": 62.44,
+      "eval_steps_per_second": 4.072,
       "step": 56
     },
     {
       "epoch": 16.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 1.7845752239227295,
-      "eval_runtime": 0.7372,
-      "eval_samples_per_second": 62.401,
-      "eval_steps_per_second": 4.07,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2858344316482544,
+      "eval_runtime": 0.7717,
+      "eval_samples_per_second": 59.607,
+      "eval_steps_per_second": 3.887,
       "step": 59
     },
     {
       "epoch": 17.14,
-      "learning_rate": 2.105263157894737e-05,
-      "loss": 2.0579,
+      "learning_rate": 0.02777777777777778,
+      "loss": 1.2597,
       "step": 60
     },
     {
       "epoch": 18.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.6324856281280518,
-      "eval_runtime": 0.7287,
-      "eval_samples_per_second": 63.129,
-      "eval_steps_per_second": 4.117,
+      "eval_loss": 1.2520229816436768,
+      "eval_runtime": 0.7474,
+      "eval_samples_per_second": 61.55,
+      "eval_steps_per_second": 4.014,
       "step": 63
     },
     {
       "epoch": 18.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.5432193279266357,
-      "eval_runtime": 0.7462,
-      "eval_samples_per_second": 61.648,
-      "eval_steps_per_second": 4.021,
+      "eval_loss": 1.2308332920074463,
+      "eval_runtime": 0.7704,
+      "eval_samples_per_second": 59.708,
+      "eval_steps_per_second": 3.894,
       "step": 66
     },
     {
       "epoch": 20.0,
-      "learning_rate": 1.754385964912281e-05,
-      "loss": 1.6272,
+      "learning_rate": 0.02314814814814815,
+      "loss": 1.2406,
       "step": 70
     },
     {
       "epoch": 20.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.4466660022735596,
-      "eval_runtime": 0.7257,
-      "eval_samples_per_second": 63.39,
-      "eval_steps_per_second": 4.134,
+      "eval_loss": 1.223983645439148,
+      "eval_runtime": 1.0122,
+      "eval_samples_per_second": 45.445,
+      "eval_steps_per_second": 2.964,
       "step": 70
     },
     {
       "epoch": 20.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.386461615562439,
-      "eval_runtime": 0.7507,
-      "eval_samples_per_second": 61.278,
-      "eval_steps_per_second": 3.996,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2277472019195557,
+      "eval_runtime": 0.7687,
+      "eval_samples_per_second": 59.844,
+      "eval_steps_per_second": 3.903,
       "step": 73
     },
     {
       "epoch": 22.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.32200026512146,
-      "eval_runtime": 0.8387,
-      "eval_samples_per_second": 54.846,
-      "eval_steps_per_second": 3.577,
+      "eval_loss": 1.243747353553772,
+      "eval_runtime": 0.7578,
+      "eval_samples_per_second": 60.702,
+      "eval_steps_per_second": 3.959,
       "step": 77
     },
     {
       "epoch": 22.86,
-      "learning_rate": 1.4035087719298246e-05,
-      "loss": 1.3699,
+      "learning_rate": 0.018518518518518517,
+      "loss": 1.2506,
       "step": 80
     },
     {
       "epoch": 22.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2847795486450195,
-      "eval_runtime": 1.0743,
-      "eval_samples_per_second": 42.82,
-      "eval_steps_per_second": 2.793,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2311103343963623,
+      "eval_runtime": 0.7393,
+      "eval_samples_per_second": 62.222,
+      "eval_steps_per_second": 4.058,
       "step": 80
     },
     {
       "epoch": 24.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2489309310913086,
-      "eval_runtime": 0.8362,
-      "eval_samples_per_second": 55.01,
-      "eval_steps_per_second": 3.588,
+      "eval_loss": 1.210280179977417,
+      "eval_runtime": 0.7435,
+      "eval_samples_per_second": 61.873,
+      "eval_steps_per_second": 4.035,
       "step": 84
     },
     {
       "epoch": 24.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2313342094421387,
-      "eval_runtime": 0.7937,
-      "eval_samples_per_second": 57.955,
-      "eval_steps_per_second": 3.78,
+      "eval_loss": 1.2071959972381592,
+      "eval_runtime": 0.8851,
+      "eval_samples_per_second": 51.974,
+      "eval_steps_per_second": 3.39,
       "step": 87
     },
     {
       "epoch": 25.71,
-      "learning_rate": 1.0526315789473684e-05,
-      "loss": 1.2469,
+      "learning_rate": 0.01388888888888889,
+      "loss": 1.2026,
       "step": 90
     },
     {
       "epoch": 26.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2175383567810059,
-      "eval_runtime": 0.7492,
-      "eval_samples_per_second": 61.401,
-      "eval_steps_per_second": 4.004,
+      "eval_loss": 1.217361569404602,
+      "eval_runtime": 0.8382,
+      "eval_samples_per_second": 54.876,
+      "eval_steps_per_second": 3.579,
       "step": 91
     },
     {
       "epoch": 26.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2125990390777588,
-      "eval_runtime": 0.7412,
-      "eval_samples_per_second": 62.063,
-      "eval_steps_per_second": 4.048,
+      "eval_loss": 1.2164480686187744,
+      "eval_runtime": 0.7529,
+      "eval_samples_per_second": 61.098,
+      "eval_steps_per_second": 3.985,
       "step": 94
     },
     {
       "epoch": 28.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2115219831466675,
-      "eval_runtime": 0.8532,
-      "eval_samples_per_second": 53.915,
-      "eval_steps_per_second": 3.516,
+      "eval_loss": 1.214943528175354,
+      "eval_runtime": 0.7901,
+      "eval_samples_per_second": 58.217,
+      "eval_steps_per_second": 3.797,
       "step": 98
     },
     {
       "epoch": 28.57,
-      "learning_rate": 7.017543859649123e-06,
-      "loss": 1.2128,
+      "learning_rate": 0.009259259259259259,
+      "loss": 1.2,
       "step": 100
     },
     {
       "epoch": 28.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2108038663864136,
-      "eval_runtime": 0.7442,
-      "eval_samples_per_second": 61.813,
-      "eval_steps_per_second": 4.031,
+      "eval_loss": 1.206203579902649,
+      "eval_runtime": 0.7505,
+      "eval_samples_per_second": 61.295,
+      "eval_steps_per_second": 3.997,
       "step": 101
     },
     {
       "epoch": 30.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2107146978378296,
-      "eval_runtime": 0.7702,
-      "eval_samples_per_second": 59.727,
-      "eval_steps_per_second": 3.895,
+      "eval_loss": 1.2065629959106445,
+      "eval_runtime": 0.7493,
+      "eval_samples_per_second": 61.394,
+      "eval_steps_per_second": 4.004,
       "step": 105
     },
     {
       "epoch": 30.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2101784944534302,
-      "eval_runtime": 0.8207,
-      "eval_samples_per_second": 56.05,
-      "eval_steps_per_second": 3.655,
+      "eval_loss": 1.208692193031311,
+      "eval_runtime": 0.7358,
+      "eval_samples_per_second": 62.513,
+      "eval_steps_per_second": 4.077,
       "step": 108
     },
     {
       "epoch": 31.43,
-      "learning_rate": 3.5087719298245615e-06,
-      "loss": 1.1889,
+      "learning_rate": 0.004629629629629629,
+      "loss": 1.2065,
       "step": 110
     },
     {
       "epoch": 32.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.209415316581726,
-      "eval_runtime": 0.8748,
-      "eval_samples_per_second": 52.586,
-      "eval_steps_per_second": 3.43,
+      "eval_loss": 1.2076929807662964,
+      "eval_runtime": 0.7508,
+      "eval_samples_per_second": 61.264,
+      "eval_steps_per_second": 3.995,
       "step": 112
     },
     {
       "epoch": 32.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.208990454673767,
-      "eval_runtime": 0.7574,
-      "eval_samples_per_second": 60.738,
-      "eval_steps_per_second": 3.961,
+      "eval_loss": 1.206972599029541,
+      "eval_runtime": 0.7519,
+      "eval_samples_per_second": 61.176,
+      "eval_steps_per_second": 3.99,
       "step": 115
     },
     {
       "epoch": 34.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2087311744689941,
-      "eval_runtime": 0.7609,
-      "eval_samples_per_second": 60.453,
-      "eval_steps_per_second": 3.943,
+      "eval_loss": 1.2067426443099976,
+      "eval_runtime": 0.7542,
+      "eval_samples_per_second": 60.994,
+      "eval_steps_per_second": 3.978,
       "step": 119
     },
     {
       "epoch": 34.29,
       "learning_rate": 0.0,
-      "loss": 1.2137,
+      "loss": 1.1998,
       "step": 120
     },
     {
       "epoch": 34.29,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.208715796470642,
-      "eval_runtime": 0.7628,
-      "eval_samples_per_second": 60.304,
-      "eval_steps_per_second": 3.933,
+      "eval_loss": 1.2067285776138306,
+      "eval_runtime": 0.8157,
+      "eval_samples_per_second": 56.395,
+      "eval_steps_per_second": 3.678,
       "step": 120
     },
     {
       "epoch": 34.29,
       "step": 120,
       "total_flos": 2.3770905934823424e+17,
-      "train_loss": 3.0923615217208864,
-      "train_runtime": 200.015,
-      "train_samples_per_second": 42.597,
+      "train_loss": 2.340296713511149,
+      "train_runtime": 200.1533,
+      "train_samples_per_second": 42.567,
       "train_steps_per_second": 0.6
     }
   ],