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: 1.2065
+- Loss: 4.2935
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -0,0 +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_results.json

@@ -0,0 +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
+}

runs/Dec02_14-13-10_DESKTOP-SKBE9FB/events.out.tfevents.1733170695.DESKTOP-SKBE9FB.15300.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c54f097475b97ea4b1ab174fbc4203114139ddbd81e0e27cae00491b1c392e2
+size 405

train_results.json

@@ -0,0 +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
+}

trainer_state.json

@@ -0,0 +1,417 @@
+{
+  "best_metric": 0.45652173913043476,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-ex\\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": 4.433082580566406,
+      "eval_runtime": 0.8144,
+      "eval_samples_per_second": 56.485,
+      "eval_steps_per_second": 3.684,
+      "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,
+      "step": 7
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.075,
+      "loss": 9.3798,
+      "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,
+      "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,
+      "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,
+      "step": 17
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 0.08333333333333333,
+      "loss": 6.904,
+      "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,
+      "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,
+      "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,
+      "step": 28
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 0.075,
+      "loss": 2.7643,
+      "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,
+      "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,
+      "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,
+      "step": 38
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 0.06666666666666667,
+      "loss": 2.4694,
+      "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,
+      "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,
+      "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,
+      "step": 49
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 0.05833333333333333,
+      "loss": 1.3638,
+      "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,
+      "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,
+      "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,
+      "step": 59
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 0.05,
+      "loss": 1.2334,
+      "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,
+      "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,
+      "step": 66
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.041666666666666664,
+      "loss": 1.2262,
+      "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,
+      "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,
+      "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,
+      "step": 77
+    },
+    {
+      "epoch": 22.86,
+      "learning_rate": 0.03333333333333333,
+      "loss": 1.2038,
+      "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,
+      "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,
+      "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,
+      "step": 87
+    },
+    {
+      "epoch": 25.71,
+      "learning_rate": 0.025,
+      "loss": 1.1975,
+      "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,
+      "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,
+      "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,
+      "step": 98
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 0.016666666666666666,
+      "loss": 1.2005,
+      "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,
+      "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,
+      "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,
+      "step": 108
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 0.008333333333333333,
+      "loss": 1.2012,
+      "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,
+      "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,
+      "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,
+      "step": 119
+    },
+    {
+      "epoch": 34.29,
+      "learning_rate": 0.0,
+      "loss": 1.1985,
+      "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,
+      "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
+    }
+  ],
+  "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
+}