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: 1.2118
+- Loss: 2.2674
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.45652173913043476,
+    "eval_loss": 2.267380714416504,
+    "eval_runtime": 0.7582,
+    "eval_samples_per_second": 60.671,
+    "eval_steps_per_second": 3.957,
+    "train_loss": 1.7081489324569703,
+    "train_runtime": 200.5493,
+    "train_samples_per_second": 42.483,
+    "train_steps_per_second": 0.598
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.45652173913043476,
+    "eval_loss": 2.267380714416504,
+    "eval_runtime": 0.7582,
+    "eval_samples_per_second": 60.671,
+    "eval_steps_per_second": 3.957
+}

runs/Dec02_16-17-51_DESKTOP-SKBE9FB/events.out.tfevents.1733178108.DESKTOP-SKBE9FB.19088.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa77d118468d849a57bed169157cd392506047159ec322a99a8c13110d82877e
+size 405

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 34.29,
+    "train_loss": 1.7081489324569703,
+    "train_runtime": 200.5493,
+    "train_samples_per_second": 42.483,
+    "train_steps_per_second": 0.598
+}

trainer_state.json

@@ -0,0 +1,417 @@
+{
+  "best_metric": 0.45652173913043476,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-4e-3\\checkpoint-21",
+  "epoch": 34.285714285714285,
+  "eval_steps": 500,
+  "global_step": 120,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 4.599056720733643,
+      "eval_runtime": 0.7447,
+      "eval_samples_per_second": 61.772,
+      "eval_steps_per_second": 4.029,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 2.3525421619415283,
+      "eval_runtime": 0.7773,
+      "eval_samples_per_second": 59.178,
+      "eval_steps_per_second": 3.859,
+      "step": 7
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.0038596491228070177,
+      "loss": 4.6082,
+      "step": 10
+    },
+    {
+      "epoch": 2.86,
+      "eval_accuracy": 0.15217391304347827,
+      "eval_loss": 2.1274752616882324,
+      "eval_runtime": 0.7397,
+      "eval_samples_per_second": 62.19,
+      "eval_steps_per_second": 4.056,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 2.1729302406311035,
+      "eval_runtime": 0.7352,
+      "eval_samples_per_second": 62.571,
+      "eval_steps_per_second": 4.081,
+      "step": 14
+    },
+    {
+      "epoch": 4.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.724708080291748,
+      "eval_runtime": 0.8052,
+      "eval_samples_per_second": 57.129,
+      "eval_steps_per_second": 3.726,
+      "step": 17
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 0.0035087719298245615,
+      "loss": 2.6299,
+      "step": 20
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.267380714416504,
+      "eval_runtime": 0.7442,
+      "eval_samples_per_second": 61.813,
+      "eval_steps_per_second": 4.031,
+      "step": 21
+    },
+    {
+      "epoch": 6.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.7187566757202148,
+      "eval_runtime": 0.7282,
+      "eval_samples_per_second": 63.171,
+      "eval_steps_per_second": 4.12,
+      "step": 24
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.3022313117980957,
+      "eval_runtime": 0.7212,
+      "eval_samples_per_second": 63.784,
+      "eval_steps_per_second": 4.16,
+      "step": 28
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 0.003157894736842105,
+      "loss": 2.0766,
+      "step": 30
+    },
+    {
+      "epoch": 8.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2425122261047363,
+      "eval_runtime": 0.7197,
+      "eval_samples_per_second": 63.918,
+      "eval_steps_per_second": 4.169,
+      "step": 31
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.285474419593811,
+      "eval_runtime": 0.7237,
+      "eval_samples_per_second": 63.565,
+      "eval_steps_per_second": 4.146,
+      "step": 35
+    },
+    {
+      "epoch": 10.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2310367822647095,
+      "eval_runtime": 0.7377,
+      "eval_samples_per_second": 62.358,
+      "eval_steps_per_second": 4.067,
+      "step": 38
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 0.002807017543859649,
+      "loss": 1.2802,
+      "step": 40
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.3592609167099,
+      "eval_runtime": 0.7472,
+      "eval_samples_per_second": 61.564,
+      "eval_steps_per_second": 4.015,
+      "step": 42
+    },
+    {
+      "epoch": 12.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2867013216018677,
+      "eval_runtime": 0.7512,
+      "eval_samples_per_second": 61.238,
+      "eval_steps_per_second": 3.994,
+      "step": 45
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2526949644088745,
+      "eval_runtime": 0.7707,
+      "eval_samples_per_second": 59.687,
+      "eval_steps_per_second": 3.893,
+      "step": 49
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 0.002456140350877193,
+      "loss": 1.3414,
+      "step": 50
+    },
+    {
+      "epoch": 14.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2312103509902954,
+      "eval_runtime": 0.7427,
+      "eval_samples_per_second": 61.939,
+      "eval_steps_per_second": 4.039,
+      "step": 52
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.211774230003357,
+      "eval_runtime": 0.8481,
+      "eval_samples_per_second": 54.238,
+      "eval_steps_per_second": 3.537,
+      "step": 56
+    },
+    {
+      "epoch": 16.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2191368341445923,
+      "eval_runtime": 0.7372,
+      "eval_samples_per_second": 62.4,
+      "eval_steps_per_second": 4.07,
+      "step": 59
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 0.002105263157894737,
+      "loss": 1.2764,
+      "step": 60
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2413936853408813,
+      "eval_runtime": 0.7582,
+      "eval_samples_per_second": 60.672,
+      "eval_steps_per_second": 3.957,
+      "step": 63
+    },
+    {
+      "epoch": 18.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2397258281707764,
+      "eval_runtime": 0.7417,
+      "eval_samples_per_second": 62.022,
+      "eval_steps_per_second": 4.045,
+      "step": 66
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0017543859649122807,
+      "loss": 1.2236,
+      "step": 70
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2237061262130737,
+      "eval_runtime": 0.7576,
+      "eval_samples_per_second": 60.72,
+      "eval_steps_per_second": 3.96,
+      "step": 70
+    },
+    {
+      "epoch": 20.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2179197072982788,
+      "eval_runtime": 0.7272,
+      "eval_samples_per_second": 63.26,
+      "eval_steps_per_second": 4.126,
+      "step": 73
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2233848571777344,
+      "eval_runtime": 0.7492,
+      "eval_samples_per_second": 61.401,
+      "eval_steps_per_second": 4.004,
+      "step": 77
+    },
+    {
+      "epoch": 22.86,
+      "learning_rate": 0.0014035087719298245,
+      "loss": 1.2294,
+      "step": 80
+    },
+    {
+      "epoch": 22.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2494986057281494,
+      "eval_runtime": 0.7792,
+      "eval_samples_per_second": 59.035,
+      "eval_steps_per_second": 3.85,
+      "step": 80
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2092479467391968,
+      "eval_runtime": 0.9052,
+      "eval_samples_per_second": 50.817,
+      "eval_steps_per_second": 3.314,
+      "step": 84
+    },
+    {
+      "epoch": 24.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2176958322525024,
+      "eval_runtime": 0.8812,
+      "eval_samples_per_second": 52.2,
+      "eval_steps_per_second": 3.404,
+      "step": 87
+    },
+    {
+      "epoch": 25.71,
+      "learning_rate": 0.0010526315789473684,
+      "loss": 1.2147,
+      "step": 90
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2105334997177124,
+      "eval_runtime": 0.8072,
+      "eval_samples_per_second": 56.989,
+      "eval_steps_per_second": 3.717,
+      "step": 91
+    },
+    {
+      "epoch": 26.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2271322011947632,
+      "eval_runtime": 0.8551,
+      "eval_samples_per_second": 53.796,
+      "eval_steps_per_second": 3.508,
+      "step": 94
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2263004779815674,
+      "eval_runtime": 0.7487,
+      "eval_samples_per_second": 61.442,
+      "eval_steps_per_second": 4.007,
+      "step": 98
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 0.0007017543859649122,
+      "loss": 1.2097,
+      "step": 100
+    },
+    {
+      "epoch": 28.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2107641696929932,
+      "eval_runtime": 0.7867,
+      "eval_samples_per_second": 58.473,
+      "eval_steps_per_second": 3.813,
+      "step": 101
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2069482803344727,
+      "eval_runtime": 0.8142,
+      "eval_samples_per_second": 56.498,
+      "eval_steps_per_second": 3.685,
+      "step": 105
+    },
+    {
+      "epoch": 30.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.209625244140625,
+      "eval_runtime": 0.7792,
+      "eval_samples_per_second": 59.037,
+      "eval_steps_per_second": 3.85,
+      "step": 108
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 0.0003508771929824561,
+      "loss": 1.2038,
+      "step": 110
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2106330394744873,
+      "eval_runtime": 0.7534,
+      "eval_samples_per_second": 61.059,
+      "eval_steps_per_second": 3.982,
+      "step": 112
+    },
+    {
+      "epoch": 32.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2102200984954834,
+      "eval_runtime": 0.7597,
+      "eval_samples_per_second": 60.551,
+      "eval_steps_per_second": 3.949,
+      "step": 115
+    },
+    {
+      "epoch": 34.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.211931824684143,
+      "eval_runtime": 0.8134,
+      "eval_samples_per_second": 56.552,
+      "eval_steps_per_second": 3.688,
+      "step": 119
+    },
+    {
+      "epoch": 34.29,
+      "learning_rate": 0.0,
+      "loss": 1.2038,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2118245363235474,
+      "eval_runtime": 0.8229,
+      "eval_samples_per_second": 55.901,
+      "eval_steps_per_second": 3.646,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "step": 120,
+      "total_flos": 2.3770905934823424e+17,
+      "train_loss": 1.7081489324569703,
+      "train_runtime": 200.5493,
+      "train_samples_per_second": 42.483,
+      "train_steps_per_second": 0.598
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 120,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 40,
+  "save_steps": 500,
+  "total_flos": 2.3770905934823424e+17,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}