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: 1.2072
+- Loss: 1.2148
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 1.3498146533966064,
-    "eval_runtime": 0.7272,
-    "eval_samples_per_second": 63.26,
-    "eval_steps_per_second": 4.126,
-    "train_loss": 1.642418885231018,
-    "train_runtime": 205.2899,
-    "train_samples_per_second": 41.502,
-    "train_steps_per_second": 0.585
+    "eval_loss": 1.2148082256317139,
+    "eval_runtime": 0.9303,
+    "eval_samples_per_second": 49.446,
+    "eval_steps_per_second": 3.225,
+    "train_loss": 1.5571933190027873,
+    "train_runtime": 205.4869,
+    "train_samples_per_second": 41.462,
+    "train_steps_per_second": 0.584
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 1.3498146533966064,
-    "eval_runtime": 0.7272,
-    "eval_samples_per_second": 63.26,
-    "eval_steps_per_second": 4.126
+    "eval_loss": 1.2148082256317139,
+    "eval_runtime": 0.9303,
+    "eval_samples_per_second": 49.446,
+    "eval_steps_per_second": 3.225
 }

runs/Dec02_14-46-00_DESKTOP-SKBE9FB/events.out.tfevents.1733172644.DESKTOP-SKBE9FB.11608.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82577dbb04ab81a606419509a189f06216ed5835d88231e22a1d1657ec7f940a
+size 405

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 34.29,
-    "train_loss": 1.642418885231018,
-    "train_runtime": 205.2899,
-    "train_samples_per_second": 41.502,
-    "train_steps_per_second": 0.585
+    "train_loss": 1.5571933190027873,
+    "train_runtime": 205.4869,
+    "train_samples_per_second": 41.462,
+    "train_steps_per_second": 0.584
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
   "best_metric": 0.45652173913043476,
-  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\checkpoint-17",
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\checkpoint-7",
   "epoch": 34.285714285714285,
   "eval_steps": 500,
   "global_step": 120,
@@ -11,398 +11,398 @@
     {
       "epoch": 0.86,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 6.1825971603393555,
-      "eval_runtime": 0.7341,
-      "eval_samples_per_second": 62.666,
-      "eval_steps_per_second": 4.087,
+      "eval_loss": 6.683231353759766,
+      "eval_runtime": 0.7677,
+      "eval_samples_per_second": 59.919,
+      "eval_steps_per_second": 3.908,
       "step": 3
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.780600666999817,
-      "eval_runtime": 0.7799,
-      "eval_samples_per_second": 58.98,
-      "eval_steps_per_second": 3.847,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2148082256317139,
+      "eval_runtime": 0.7682,
+      "eval_samples_per_second": 59.881,
+      "eval_steps_per_second": 3.905,
       "step": 7
     },
     {
       "epoch": 2.86,
-      "learning_rate": 0.004166666666666667,
-      "loss": 4.802,
+      "learning_rate": 0.0033333333333333335,
+      "loss": 4.4686,
       "step": 10
     },
     {
       "epoch": 2.86,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 2.233830690383911,
-      "eval_runtime": 0.7934,
-      "eval_samples_per_second": 57.978,
-      "eval_steps_per_second": 3.781,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 2.5061397552490234,
+      "eval_runtime": 0.8667,
+      "eval_samples_per_second": 53.075,
+      "eval_steps_per_second": 3.461,
       "step": 10
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.4061155319213867,
-      "eval_runtime": 0.7488,
-      "eval_samples_per_second": 61.429,
-      "eval_steps_per_second": 4.006,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.4142353534698486,
+      "eval_runtime": 0.7492,
+      "eval_samples_per_second": 61.401,
+      "eval_steps_per_second": 4.004,
       "step": 14
     },
     {
       "epoch": 4.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.3498146533966064,
-      "eval_runtime": 0.7432,
-      "eval_samples_per_second": 61.896,
-      "eval_steps_per_second": 4.037,
+      "eval_loss": 1.6117892265319824,
+      "eval_runtime": 0.8195,
+      "eval_samples_per_second": 56.133,
+      "eval_steps_per_second": 3.661,
       "step": 17
     },
     {
       "epoch": 5.71,
-      "learning_rate": 0.004629629629629629,
-      "loss": 2.3987,
+      "learning_rate": 0.003703703703703704,
+      "loss": 1.7414,
       "step": 20
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 1.5674153566360474,
-      "eval_runtime": 0.8123,
-      "eval_samples_per_second": 56.632,
-      "eval_steps_per_second": 3.693,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2483736276626587,
+      "eval_runtime": 0.7888,
+      "eval_samples_per_second": 58.319,
+      "eval_steps_per_second": 3.803,
       "step": 21
     },
     {
       "epoch": 6.86,
       "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.3750334978103638,
-      "eval_runtime": 0.749,
-      "eval_samples_per_second": 61.412,
-      "eval_steps_per_second": 4.005,
+      "eval_loss": 1.3689510822296143,
+      "eval_runtime": 0.8157,
+      "eval_samples_per_second": 56.393,
+      "eval_steps_per_second": 3.678,
       "step": 24
     },
     {
       "epoch": 8.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.4330613613128662,
-      "eval_runtime": 0.8423,
-      "eval_samples_per_second": 54.614,
-      "eval_steps_per_second": 3.562,
+      "eval_loss": 1.4064823389053345,
+      "eval_runtime": 0.7631,
+      "eval_samples_per_second": 60.277,
+      "eval_steps_per_second": 3.931,
       "step": 28
     },
     {
       "epoch": 8.57,
-      "learning_rate": 0.004166666666666667,
-      "loss": 1.4401,
+      "learning_rate": 0.0033333333333333335,
+      "loss": 1.3568,
       "step": 30
     },
     {
       "epoch": 8.86,
       "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.278549313545227,
-      "eval_runtime": 0.7572,
-      "eval_samples_per_second": 60.752,
-      "eval_steps_per_second": 3.962,
+      "eval_loss": 1.2682174444198608,
+      "eval_runtime": 0.8441,
+      "eval_samples_per_second": 54.498,
+      "eval_steps_per_second": 3.554,
       "step": 31
     },
     {
       "epoch": 10.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2276438474655151,
-      "eval_runtime": 0.7458,
-      "eval_samples_per_second": 61.681,
-      "eval_steps_per_second": 4.023,
+      "eval_loss": 1.2140024900436401,
+      "eval_runtime": 0.7452,
+      "eval_samples_per_second": 61.731,
+      "eval_steps_per_second": 4.026,
       "step": 35
     },
     {
       "epoch": 10.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2260234355926514,
-      "eval_runtime": 0.7684,
-      "eval_samples_per_second": 59.867,
-      "eval_steps_per_second": 3.904,
+      "eval_loss": 1.2590564489364624,
+      "eval_runtime": 0.8142,
+      "eval_samples_per_second": 56.494,
+      "eval_steps_per_second": 3.684,
       "step": 38
     },
     {
       "epoch": 11.43,
-      "learning_rate": 0.0037037037037037034,
-      "loss": 1.2542,
+      "learning_rate": 0.002962962962962963,
+      "loss": 1.2275,
       "step": 40
     },
     {
       "epoch": 12.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.222497582435608,
-      "eval_runtime": 0.7642,
-      "eval_samples_per_second": 60.196,
-      "eval_steps_per_second": 3.926,
+      "eval_loss": 1.2518616914749146,
+      "eval_runtime": 0.7487,
+      "eval_samples_per_second": 61.44,
+      "eval_steps_per_second": 4.007,
       "step": 42
     },
     {
       "epoch": 12.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2367029190063477,
-      "eval_runtime": 0.7662,
-      "eval_samples_per_second": 60.037,
-      "eval_steps_per_second": 3.915,
+      "eval_loss": 1.2184066772460938,
+      "eval_runtime": 0.8182,
+      "eval_samples_per_second": 56.219,
+      "eval_steps_per_second": 3.666,
       "step": 45
     },
     {
       "epoch": 14.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2583808898925781,
-      "eval_runtime": 0.7507,
-      "eval_samples_per_second": 61.278,
-      "eval_steps_per_second": 3.996,
+      "eval_loss": 1.2592365741729736,
+      "eval_runtime": 0.7547,
+      "eval_samples_per_second": 60.953,
+      "eval_steps_per_second": 3.975,
       "step": 49
     },
     {
       "epoch": 14.29,
-      "learning_rate": 0.0032407407407407406,
-      "loss": 1.3069,
+      "learning_rate": 0.0025925925925925925,
+      "loss": 1.3025,
       "step": 50
     },
     {
       "epoch": 14.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2061717510223389,
-      "eval_runtime": 0.7671,
-      "eval_samples_per_second": 59.966,
-      "eval_steps_per_second": 3.911,
+      "eval_loss": 1.2246149778366089,
+      "eval_runtime": 0.7633,
+      "eval_samples_per_second": 60.267,
+      "eval_steps_per_second": 3.93,
       "step": 52
     },
     {
       "epoch": 16.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2094433307647705,
-      "eval_runtime": 0.7677,
-      "eval_samples_per_second": 59.917,
-      "eval_steps_per_second": 3.908,
+      "eval_loss": 1.3045893907546997,
+      "eval_runtime": 0.7571,
+      "eval_samples_per_second": 60.762,
+      "eval_steps_per_second": 3.963,
       "step": 56
     },
     {
       "epoch": 16.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.235077142715454,
-      "eval_runtime": 0.7767,
-      "eval_samples_per_second": 59.223,
-      "eval_steps_per_second": 3.862,
+      "eval_loss": 1.2177391052246094,
+      "eval_runtime": 0.7723,
+      "eval_samples_per_second": 59.56,
+      "eval_steps_per_second": 3.884,
       "step": 59
     },
     {
       "epoch": 17.14,
-      "learning_rate": 0.002777777777777778,
-      "loss": 1.2418,
+      "learning_rate": 0.0022222222222222222,
+      "loss": 1.2981,
       "step": 60
     },
     {
       "epoch": 18.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2273476123809814,
-      "eval_runtime": 0.7532,
-      "eval_samples_per_second": 61.073,
-      "eval_steps_per_second": 3.983,
+      "eval_loss": 1.2338941097259521,
+      "eval_runtime": 0.7731,
+      "eval_samples_per_second": 59.498,
+      "eval_steps_per_second": 3.88,
       "step": 63
     },
     {
       "epoch": 18.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2196640968322754,
-      "eval_runtime": 0.9389,
-      "eval_samples_per_second": 48.993,
-      "eval_steps_per_second": 3.195,
+      "eval_loss": 1.313918113708496,
+      "eval_runtime": 0.7563,
+      "eval_samples_per_second": 60.821,
+      "eval_steps_per_second": 3.967,
       "step": 66
     },
     {
       "epoch": 20.0,
-      "learning_rate": 0.0023148148148148147,
-      "loss": 1.2399,
+      "learning_rate": 0.001851851851851852,
+      "loss": 1.2765,
       "step": 70
     },
     {
       "epoch": 20.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.233923316001892,
-      "eval_runtime": 0.8619,
-      "eval_samples_per_second": 53.373,
-      "eval_steps_per_second": 3.481,
+      "eval_loss": 1.2116084098815918,
+      "eval_runtime": 0.7487,
+      "eval_samples_per_second": 61.441,
+      "eval_steps_per_second": 4.007,
       "step": 70
     },
     {
       "epoch": 20.86,
       "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2420306205749512,
-      "eval_runtime": 0.7943,
-      "eval_samples_per_second": 57.911,
-      "eval_steps_per_second": 3.777,
+      "eval_loss": 1.2283920049667358,
+      "eval_runtime": 0.8009,
+      "eval_samples_per_second": 57.432,
+      "eval_steps_per_second": 3.746,
       "step": 73
     },
     {
       "epoch": 22.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.230379581451416,
-      "eval_runtime": 0.7612,
-      "eval_samples_per_second": 60.433,
-      "eval_steps_per_second": 3.941,
+      "eval_loss": 1.2246237993240356,
+      "eval_runtime": 0.7673,
+      "eval_samples_per_second": 59.952,
+      "eval_steps_per_second": 3.91,
       "step": 77
     },
     {
       "epoch": 22.86,
-      "learning_rate": 0.0018518518518518517,
-      "loss": 1.2262,
+      "learning_rate": 0.0014814814814814814,
+      "loss": 1.2074,
       "step": 80
     },
     {
       "epoch": 22.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2175580263137817,
-      "eval_runtime": 0.8099,
-      "eval_samples_per_second": 56.794,
-      "eval_steps_per_second": 3.704,
+      "eval_loss": 1.2630535364151,
+      "eval_runtime": 0.7811,
+      "eval_samples_per_second": 58.89,
+      "eval_steps_per_second": 3.841,
       "step": 80
     },
     {
       "epoch": 24.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2133561372756958,
-      "eval_runtime": 0.7651,
-      "eval_samples_per_second": 60.124,
-      "eval_steps_per_second": 3.921,
+      "eval_loss": 1.2092362642288208,
+      "eval_runtime": 0.769,
+      "eval_samples_per_second": 59.821,
+      "eval_steps_per_second": 3.901,
       "step": 84
     },
     {
       "epoch": 24.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2082937955856323,
-      "eval_runtime": 0.7574,
-      "eval_samples_per_second": 60.736,
-      "eval_steps_per_second": 3.961,
+      "eval_loss": 1.2146543264389038,
+      "eval_runtime": 0.7867,
+      "eval_samples_per_second": 58.472,
+      "eval_steps_per_second": 3.813,
       "step": 87
     },
     {
       "epoch": 25.71,
-      "learning_rate": 0.001388888888888889,
-      "loss": 1.1989,
+      "learning_rate": 0.0011111111111111111,
+      "loss": 1.2048,
       "step": 90
     },
     {
       "epoch": 26.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2147068977355957,
-      "eval_runtime": 0.7509,
-      "eval_samples_per_second": 61.261,
-      "eval_steps_per_second": 3.995,
+      "eval_loss": 1.2120637893676758,
+      "eval_runtime": 0.8352,
+      "eval_samples_per_second": 55.077,
+      "eval_steps_per_second": 3.592,
       "step": 91
     },
     {
       "epoch": 26.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2231361865997314,
-      "eval_runtime": 0.7897,
-      "eval_samples_per_second": 58.25,
-      "eval_steps_per_second": 3.799,
+      "eval_loss": 1.215645432472229,
+      "eval_runtime": 0.7342,
+      "eval_samples_per_second": 62.656,
+      "eval_steps_per_second": 4.086,
       "step": 94
     },
     {
       "epoch": 28.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2130694389343262,
-      "eval_runtime": 0.7517,
-      "eval_samples_per_second": 61.192,
-      "eval_steps_per_second": 3.991,
+      "eval_loss": 1.2249161005020142,
+      "eval_runtime": 0.8011,
+      "eval_samples_per_second": 57.422,
+      "eval_steps_per_second": 3.745,
       "step": 98
     },
     {
       "epoch": 28.57,
-      "learning_rate": 0.0009259259259259259,
-      "loss": 1.1991,
+      "learning_rate": 0.0007407407407407407,
+      "loss": 1.2068,
       "step": 100
     },
     {
       "epoch": 28.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2088346481323242,
-      "eval_runtime": 0.7835,
-      "eval_samples_per_second": 58.71,
-      "eval_steps_per_second": 3.829,
+      "eval_loss": 1.2159205675125122,
+      "eval_runtime": 0.7602,
+      "eval_samples_per_second": 60.509,
+      "eval_steps_per_second": 3.946,
       "step": 101
     },
     {
       "epoch": 30.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2072715759277344,
-      "eval_runtime": 0.751,
-      "eval_samples_per_second": 61.253,
-      "eval_steps_per_second": 3.995,
+      "eval_loss": 1.210811972618103,
+      "eval_runtime": 0.795,
+      "eval_samples_per_second": 57.858,
+      "eval_steps_per_second": 3.773,
       "step": 105
     },
     {
       "epoch": 30.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2108794450759888,
-      "eval_runtime": 0.7602,
-      "eval_samples_per_second": 60.513,
-      "eval_steps_per_second": 3.946,
+      "eval_loss": 1.2115787267684937,
+      "eval_runtime": 0.7682,
+      "eval_samples_per_second": 59.882,
+      "eval_steps_per_second": 3.905,
       "step": 108
     },
     {
       "epoch": 31.43,
-      "learning_rate": 0.0004629629629629629,
-      "loss": 1.2015,
+      "learning_rate": 0.00037037037037037035,
+      "loss": 1.1961,
       "step": 110
     },
     {
       "epoch": 32.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2092784643173218,
-      "eval_runtime": 1.4373,
-      "eval_samples_per_second": 32.004,
-      "eval_steps_per_second": 2.087,
+      "eval_loss": 1.2078149318695068,
+      "eval_runtime": 0.8022,
+      "eval_samples_per_second": 57.342,
+      "eval_steps_per_second": 3.74,
       "step": 112
     },
     {
       "epoch": 32.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2075750827789307,
-      "eval_runtime": 0.7502,
-      "eval_samples_per_second": 61.318,
-      "eval_steps_per_second": 3.999,
+      "eval_loss": 1.2069722414016724,
+      "eval_runtime": 0.8538,
+      "eval_samples_per_second": 53.877,
+      "eval_steps_per_second": 3.514,
       "step": 115
     },
     {
       "epoch": 34.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2066638469696045,
-      "eval_runtime": 0.7312,
-      "eval_samples_per_second": 62.913,
-      "eval_steps_per_second": 4.103,
+      "eval_loss": 1.2071927785873413,
+      "eval_runtime": 0.77,
+      "eval_samples_per_second": 59.74,
+      "eval_steps_per_second": 3.896,
       "step": 119
     },
     {
       "epoch": 34.29,
       "learning_rate": 0.0,
-      "loss": 1.1997,
+      "loss": 1.1999,
       "step": 120
     },
     {
       "epoch": 34.29,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2064862251281738,
-      "eval_runtime": 0.7487,
-      "eval_samples_per_second": 61.442,
-      "eval_steps_per_second": 4.007,
+      "eval_loss": 1.207156777381897,
+      "eval_runtime": 0.7739,
+      "eval_samples_per_second": 59.437,
+      "eval_steps_per_second": 3.876,
       "step": 120
     },
     {
       "epoch": 34.29,
       "step": 120,
       "total_flos": 2.3770905934823424e+17,
-      "train_loss": 1.642418885231018,
-      "train_runtime": 205.2899,
-      "train_samples_per_second": 41.502,
-      "train_steps_per_second": 0.585
+      "train_loss": 1.5571933190027873,
+      "train_runtime": 205.4869,
+      "train_samples_per_second": 41.462,
+      "train_steps_per_second": 0.584
     }
   ],
   "logging_steps": 10,