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: 11.3982
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 1.3785392045974731,
-    "eval_runtime": 0.9635,
-    "eval_samples_per_second": 47.74,
-    "eval_steps_per_second": 3.113,
-    "train_loss": 2.3613997220993044,
-    "train_runtime": 205.5721,
-    "train_samples_per_second": 41.445,
-    "train_steps_per_second": 0.584
+    "eval_loss": 11.398162841796875,
+    "eval_runtime": 1.0913,
+    "eval_samples_per_second": 42.151,
+    "eval_steps_per_second": 2.749,
+    "train_loss": 5.366930524508159,
+    "train_runtime": 206.997,
+    "train_samples_per_second": 41.16,
+    "train_steps_per_second": 0.58
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 34.29,
     "eval_accuracy": 0.45652173913043476,
-    "eval_loss": 1.3785392045974731,
-    "eval_runtime": 0.9635,
-    "eval_samples_per_second": 47.74,
-    "eval_steps_per_second": 3.113
+    "eval_loss": 11.398162841796875,
+    "eval_runtime": 1.0913,
+    "eval_samples_per_second": 42.151,
+    "eval_steps_per_second": 2.749
 }

runs/Dec02_15-16-55_DESKTOP-SKBE9FB/events.out.tfevents.1733174487.DESKTOP-SKBE9FB.11568.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0575a4847792733321c9558653c971701a3b596163c3225148e33fcc4af74d2
+size 405

train_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 34.29,
-    "train_loss": 2.3613997220993044,
-    "train_runtime": 205.5721,
-    "train_samples_per_second": 41.445,
-    "train_steps_per_second": 0.584
+    "train_loss": 5.366930524508159,
+    "train_runtime": 206.997,
+    "train_samples_per_second": 41.16,
+    "train_steps_per_second": 0.58
 }

trainer_state.json

@@ -10,399 +10,399 @@
   "log_history": [
     {
       "epoch": 0.86,
-      "eval_accuracy": 0.17391304347826086,
-      "eval_loss": 26.20158576965332,
-      "eval_runtime": 0.8072,
-      "eval_samples_per_second": 56.986,
-      "eval_steps_per_second": 3.716,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 45.53202438354492,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 60.431,
+      "eval_steps_per_second": 3.941,
       "step": 3
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.3785392045974731,
-      "eval_runtime": 0.7527,
-      "eval_samples_per_second": 61.111,
-      "eval_steps_per_second": 3.986,
+      "eval_loss": 11.398162841796875,
+      "eval_runtime": 0.7452,
+      "eval_samples_per_second": 61.727,
+      "eval_steps_per_second": 4.026,
       "step": 7
     },
     {
       "epoch": 2.86,
-      "learning_rate": 0.018333333333333333,
-      "loss": 12.975,
+      "learning_rate": 0.09166666666666667,
+      "loss": 37.3882,
       "step": 10
     },
     {
       "epoch": 2.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 2.2855052947998047,
-      "eval_runtime": 0.8039,
-      "eval_samples_per_second": 57.219,
-      "eval_steps_per_second": 3.732,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 14.659239768981934,
+      "eval_runtime": 0.7983,
+      "eval_samples_per_second": 57.624,
+      "eval_steps_per_second": 3.758,
       "step": 10
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.5437300205230713,
-      "eval_runtime": 0.7595,
-      "eval_samples_per_second": 60.57,
-      "eval_steps_per_second": 3.95,
+      "eval_loss": 5.432069301605225,
+      "eval_runtime": 0.8472,
+      "eval_samples_per_second": 54.298,
+      "eval_steps_per_second": 3.541,
       "step": 14
     },
     {
       "epoch": 4.86,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.5017277002334595,
-      "eval_runtime": 1.3666,
-      "eval_samples_per_second": 33.66,
-      "eval_steps_per_second": 2.195,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 2.191324234008789,
+      "eval_runtime": 0.7984,
+      "eval_samples_per_second": 57.617,
+      "eval_steps_per_second": 3.758,
       "step": 17
     },
     {
       "epoch": 5.71,
-      "learning_rate": 0.016666666666666666,
-      "loss": 2.1282,
+      "learning_rate": 0.08333333333333334,
+      "loss": 7.8109,
       "step": 20
     },
     {
       "epoch": 6.0,
       "eval_accuracy": 0.10869565217391304,
-      "eval_loss": 1.5409127473831177,
-      "eval_runtime": 0.8412,
-      "eval_samples_per_second": 54.684,
-      "eval_steps_per_second": 3.566,
+      "eval_loss": 7.573766231536865,
+      "eval_runtime": 0.8305,
+      "eval_samples_per_second": 55.388,
+      "eval_steps_per_second": 3.612,
       "step": 21
     },
     {
       "epoch": 6.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.4040182828903198,
-      "eval_runtime": 1.1106,
-      "eval_samples_per_second": 41.42,
-      "eval_steps_per_second": 2.701,
+      "eval_loss": 8.57021427154541,
+      "eval_runtime": 0.7898,
+      "eval_samples_per_second": 58.244,
+      "eval_steps_per_second": 3.799,
       "step": 24
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.27801513671875,
-      "eval_runtime": 0.7688,
-      "eval_samples_per_second": 59.837,
-      "eval_steps_per_second": 3.902,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 5.530113220214844,
+      "eval_runtime": 0.7503,
+      "eval_samples_per_second": 61.31,
+      "eval_steps_per_second": 3.998,
       "step": 28
     },
     {
       "epoch": 8.57,
-      "learning_rate": 0.015,
-      "loss": 1.554,
+      "learning_rate": 0.07500000000000001,
+      "loss": 6.7711,
       "step": 30
     },
     {
       "epoch": 8.86,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2299892902374268,
-      "eval_runtime": 0.7581,
-      "eval_samples_per_second": 60.68,
-      "eval_steps_per_second": 3.957,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.6876065731048584,
+      "eval_runtime": 0.8036,
+      "eval_samples_per_second": 57.245,
+      "eval_steps_per_second": 3.733,
       "step": 31
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.3227636814117432,
-      "eval_runtime": 0.7707,
-      "eval_samples_per_second": 59.684,
-      "eval_steps_per_second": 3.892,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1.8742014169692993,
+      "eval_runtime": 0.864,
+      "eval_samples_per_second": 53.242,
+      "eval_steps_per_second": 3.472,
       "step": 35
     },
     {
       "epoch": 10.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2744935750961304,
-      "eval_runtime": 0.7587,
-      "eval_samples_per_second": 60.626,
-      "eval_steps_per_second": 3.954,
+      "eval_loss": 1.5266066789627075,
+      "eval_runtime": 0.8225,
+      "eval_samples_per_second": 55.927,
+      "eval_steps_per_second": 3.647,
       "step": 38
     },
     {
       "epoch": 11.43,
-      "learning_rate": 0.013333333333333332,
-      "loss": 1.3748,
+      "learning_rate": 0.06666666666666667,
+      "loss": 1.7995,
       "step": 40
     },
     {
       "epoch": 12.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.3723951578140259,
-      "eval_runtime": 0.7682,
-      "eval_samples_per_second": 59.881,
-      "eval_steps_per_second": 3.905,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1.5311237573623657,
+      "eval_runtime": 0.7469,
+      "eval_samples_per_second": 61.585,
+      "eval_steps_per_second": 4.016,
       "step": 42
     },
     {
       "epoch": 12.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.3726118803024292,
-      "eval_runtime": 0.7647,
-      "eval_samples_per_second": 60.155,
-      "eval_steps_per_second": 3.923,
+      "eval_loss": 1.4438916444778442,
+      "eval_runtime": 0.8487,
+      "eval_samples_per_second": 54.199,
+      "eval_steps_per_second": 3.535,
       "step": 45
     },
     {
       "epoch": 14.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.289075493812561,
-      "eval_runtime": 0.7652,
-      "eval_samples_per_second": 60.115,
-      "eval_steps_per_second": 3.921,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2218061685562134,
+      "eval_runtime": 0.7762,
+      "eval_samples_per_second": 59.266,
+      "eval_steps_per_second": 3.865,
       "step": 49
     },
     {
       "epoch": 14.29,
-      "learning_rate": 0.011666666666666667,
-      "loss": 1.5315,
+      "learning_rate": 0.05833333333333334,
+      "loss": 1.5366,
       "step": 50
     },
     {
       "epoch": 14.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2978594303131104,
-      "eval_runtime": 0.7573,
-      "eval_samples_per_second": 60.741,
-      "eval_steps_per_second": 3.961,
+      "eval_loss": 1.322609305381775,
+      "eval_runtime": 0.7716,
+      "eval_samples_per_second": 59.618,
+      "eval_steps_per_second": 3.888,
       "step": 52
     },
     {
       "epoch": 16.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2271596193313599,
-      "eval_runtime": 0.8407,
-      "eval_samples_per_second": 54.715,
-      "eval_steps_per_second": 3.568,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1.6288353204727173,
+      "eval_runtime": 0.8457,
+      "eval_samples_per_second": 54.39,
+      "eval_steps_per_second": 3.547,
       "step": 56
     },
     {
       "epoch": 16.86,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2749192714691162,
-      "eval_runtime": 0.7412,
-      "eval_samples_per_second": 62.063,
-      "eval_steps_per_second": 4.048,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.7525928020477295,
+      "eval_runtime": 0.8164,
+      "eval_samples_per_second": 56.343,
+      "eval_steps_per_second": 3.675,
       "step": 59
     },
     {
       "epoch": 17.14,
-      "learning_rate": 0.01,
-      "loss": 1.351,
+      "learning_rate": 0.05,
+      "loss": 1.5748,
       "step": 60
     },
     {
       "epoch": 18.0,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2218742370605469,
-      "eval_runtime": 0.7782,
-      "eval_samples_per_second": 59.109,
-      "eval_steps_per_second": 3.855,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.369892954826355,
+      "eval_runtime": 0.817,
+      "eval_samples_per_second": 56.302,
+      "eval_steps_per_second": 3.672,
       "step": 63
     },
     {
       "epoch": 18.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2199996709823608,
-      "eval_runtime": 0.7592,
-      "eval_samples_per_second": 60.591,
-      "eval_steps_per_second": 3.952,
+      "eval_loss": 1.2662850618362427,
+      "eval_runtime": 0.7845,
+      "eval_samples_per_second": 58.632,
+      "eval_steps_per_second": 3.824,
       "step": 66
     },
     {
       "epoch": 20.0,
-      "learning_rate": 0.008333333333333333,
-      "loss": 1.2678,
+      "learning_rate": 0.04166666666666667,
+      "loss": 1.3933,
       "step": 70
     },
     {
       "epoch": 20.0,
-      "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2277557849884033,
-      "eval_runtime": 0.7951,
-      "eval_samples_per_second": 57.858,
-      "eval_steps_per_second": 3.773,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2221952676773071,
+      "eval_runtime": 0.8295,
+      "eval_samples_per_second": 55.453,
+      "eval_steps_per_second": 3.617,
       "step": 70
     },
     {
       "epoch": 20.86,
-      "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2318228483200073,
-      "eval_runtime": 0.7641,
-      "eval_samples_per_second": 60.201,
-      "eval_steps_per_second": 3.926,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.238777756690979,
+      "eval_runtime": 1.2232,
+      "eval_samples_per_second": 37.606,
+      "eval_steps_per_second": 2.453,
       "step": 73
     },
     {
       "epoch": 22.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.210179328918457,
-      "eval_runtime": 0.8337,
-      "eval_samples_per_second": 55.176,
-      "eval_steps_per_second": 3.598,
+      "eval_loss": 1.2831048965454102,
+      "eval_runtime": 0.7467,
+      "eval_samples_per_second": 61.605,
+      "eval_steps_per_second": 4.018,
       "step": 77
     },
     {
       "epoch": 22.86,
-      "learning_rate": 0.006666666666666666,
-      "loss": 1.244,
+      "learning_rate": 0.03333333333333333,
+      "loss": 1.2788,
       "step": 80
     },
     {
       "epoch": 22.86,
       "eval_accuracy": 0.32608695652173914,
-      "eval_loss": 1.2466037273406982,
-      "eval_runtime": 0.7413,
-      "eval_samples_per_second": 62.052,
-      "eval_steps_per_second": 4.047,
+      "eval_loss": 1.2515243291854858,
+      "eval_runtime": 0.768,
+      "eval_samples_per_second": 59.895,
+      "eval_steps_per_second": 3.906,
       "step": 80
     },
     {
       "epoch": 24.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2102508544921875,
-      "eval_runtime": 0.7646,
-      "eval_samples_per_second": 60.165,
-      "eval_steps_per_second": 3.924,
+      "eval_loss": 1.2105132341384888,
+      "eval_runtime": 0.7735,
+      "eval_samples_per_second": 59.469,
+      "eval_steps_per_second": 3.878,
       "step": 84
     },
     {
       "epoch": 24.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2067211866378784,
-      "eval_runtime": 0.7768,
-      "eval_samples_per_second": 59.217,
-      "eval_steps_per_second": 3.862,
+      "eval_loss": 1.214127779006958,
+      "eval_runtime": 0.7825,
+      "eval_samples_per_second": 58.787,
+      "eval_steps_per_second": 3.834,
       "step": 87
     },
     {
       "epoch": 25.71,
-      "learning_rate": 0.005,
-      "loss": 1.2585,
+      "learning_rate": 0.025,
+      "loss": 1.2218,
       "step": 90
     },
     {
       "epoch": 26.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.212925910949707,
-      "eval_runtime": 0.8062,
-      "eval_samples_per_second": 57.056,
-      "eval_steps_per_second": 3.721,
+      "eval_loss": 1.2214648723602295,
+      "eval_runtime": 0.7783,
+      "eval_samples_per_second": 59.1,
+      "eval_steps_per_second": 3.854,
       "step": 91
     },
     {
       "epoch": 26.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2109757661819458,
-      "eval_runtime": 0.7677,
-      "eval_samples_per_second": 59.921,
-      "eval_steps_per_second": 3.908,
+      "eval_loss": 1.2189139127731323,
+      "eval_runtime": 0.779,
+      "eval_samples_per_second": 59.048,
+      "eval_steps_per_second": 3.851,
       "step": 94
     },
     {
       "epoch": 28.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.213108777999878,
-      "eval_runtime": 0.8052,
-      "eval_samples_per_second": 57.129,
-      "eval_steps_per_second": 3.726,
+      "eval_loss": 1.210168719291687,
+      "eval_runtime": 0.7611,
+      "eval_samples_per_second": 60.441,
+      "eval_steps_per_second": 3.942,
       "step": 98
     },
     {
       "epoch": 28.57,
-      "learning_rate": 0.003333333333333333,
-      "loss": 1.2405,
+      "learning_rate": 0.016666666666666666,
+      "loss": 1.2039,
       "step": 100
     },
     {
       "epoch": 28.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2072187662124634,
-      "eval_runtime": 0.7726,
-      "eval_samples_per_second": 59.536,
-      "eval_steps_per_second": 3.883,
+      "eval_loss": 1.2094049453735352,
+      "eval_runtime": 0.7482,
+      "eval_samples_per_second": 61.482,
+      "eval_steps_per_second": 4.01,
       "step": 101
     },
     {
       "epoch": 30.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.209886908531189,
-      "eval_runtime": 0.8687,
-      "eval_samples_per_second": 52.953,
-      "eval_steps_per_second": 3.453,
+      "eval_loss": 1.206476092338562,
+      "eval_runtime": 0.7612,
+      "eval_samples_per_second": 60.431,
+      "eval_steps_per_second": 3.941,
       "step": 105
     },
     {
       "epoch": 30.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.211473822593689,
-      "eval_runtime": 0.7408,
-      "eval_samples_per_second": 62.098,
-      "eval_steps_per_second": 4.05,
+      "eval_loss": 1.2124706506729126,
+      "eval_runtime": 0.761,
+      "eval_samples_per_second": 60.451,
+      "eval_steps_per_second": 3.942,
       "step": 108
     },
     {
       "epoch": 31.43,
-      "learning_rate": 0.0016666666666666666,
-      "loss": 1.2134,
+      "learning_rate": 0.008333333333333333,
+      "loss": 1.2131,
       "step": 110
     },
     {
       "epoch": 32.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2137932777404785,
-      "eval_runtime": 0.7962,
-      "eval_samples_per_second": 57.776,
-      "eval_steps_per_second": 3.768,
+      "eval_loss": 1.2106635570526123,
+      "eval_runtime": 0.8627,
+      "eval_samples_per_second": 53.321,
+      "eval_steps_per_second": 3.477,
       "step": 112
     },
     {
       "epoch": 32.86,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2094972133636475,
-      "eval_runtime": 0.7562,
-      "eval_samples_per_second": 60.832,
-      "eval_steps_per_second": 3.967,
+      "eval_loss": 1.2078120708465576,
+      "eval_runtime": 0.7667,
+      "eval_samples_per_second": 59.998,
+      "eval_steps_per_second": 3.913,
       "step": 115
     },
     {
       "epoch": 34.0,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2080820798873901,
-      "eval_runtime": 0.7958,
-      "eval_samples_per_second": 57.802,
-      "eval_steps_per_second": 3.77,
+      "eval_loss": 1.2067822217941284,
+      "eval_runtime": 0.7636,
+      "eval_samples_per_second": 60.24,
+      "eval_steps_per_second": 3.929,
       "step": 119
     },
     {
       "epoch": 34.29,
       "learning_rate": 0.0,
-      "loss": 1.1982,
+      "loss": 1.211,
       "step": 120
     },
     {
       "epoch": 34.29,
       "eval_accuracy": 0.45652173913043476,
-      "eval_loss": 1.2080086469650269,
-      "eval_runtime": 0.8938,
-      "eval_samples_per_second": 51.468,
-      "eval_steps_per_second": 3.357,
+      "eval_loss": 1.2067316770553589,
+      "eval_runtime": 0.9703,
+      "eval_samples_per_second": 47.408,
+      "eval_steps_per_second": 3.092,
       "step": 120
     },
     {
       "epoch": 34.29,
       "step": 120,
       "total_flos": 2.3770905934823424e+17,
-      "train_loss": 2.3613997220993044,
-      "train_runtime": 205.5721,
-      "train_samples_per_second": 41.445,
-      "train_steps_per_second": 0.584
+      "train_loss": 5.366930524508159,
+      "train_runtime": 206.997,
+      "train_samples_per_second": 41.16,
+      "train_steps_per_second": 0.58
     }
   ],
   "logging_steps": 10,