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: 1.2338
+- Loss: 54.3349
 - Accuracy: 0.4565
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.45652173913043476,
+    "eval_loss": 54.33485412597656,
+    "eval_runtime": 1.0835,
+    "eval_samples_per_second": 42.455,
+    "eval_steps_per_second": 2.769,
+    "train_loss": 78.35383641719818,
+    "train_runtime": 207.9228,
+    "train_samples_per_second": 40.977,
+    "train_steps_per_second": 0.577
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.45652173913043476,
+    "eval_loss": 54.33485412597656,
+    "eval_runtime": 1.0835,
+    "eval_samples_per_second": 42.455,
+    "eval_steps_per_second": 2.769
+}

runs/Dec02_15-23-48_DESKTOP-SKBE9FB/events.out.tfevents.1733174883.DESKTOP-SKBE9FB.6524.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e213fb976c2cba97abafb8e4e93f5a8b5786ca3866e1fa8dc1d668ba3fca1a3
+size 405

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 34.29,
+    "train_loss": 78.35383641719818,
+    "train_runtime": 207.9228,
+    "train_samples_per_second": 40.977,
+    "train_steps_per_second": 0.577
+}

trainer_state.json

@@ -0,0 +1,417 @@
+{
+  "best_metric": 0.45652173913043476,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-5e-1\\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": 0.32608695652173914,
+      "eval_loss": 343.8003234863281,
+      "eval_runtime": 0.7367,
+      "eval_samples_per_second": 62.439,
+      "eval_steps_per_second": 4.072,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 1260.14892578125,
+      "eval_runtime": 0.8432,
+      "eval_samples_per_second": 54.555,
+      "eval_steps_per_second": 3.558,
+      "step": 7
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.4583333333333333,
+      "loss": 517.6714,
+      "step": 10
+    },
+    {
+      "epoch": 2.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 457.51031494140625,
+      "eval_runtime": 0.7677,
+      "eval_samples_per_second": 59.921,
+      "eval_steps_per_second": 3.908,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 180.12049865722656,
+      "eval_runtime": 0.8105,
+      "eval_samples_per_second": 56.752,
+      "eval_steps_per_second": 3.701,
+      "step": 14
+    },
+    {
+      "epoch": 4.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 190.96266174316406,
+      "eval_runtime": 0.7816,
+      "eval_samples_per_second": 58.851,
+      "eval_steps_per_second": 3.838,
+      "step": 17
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 0.4166666666666667,
+      "loss": 201.7487,
+      "step": 20
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 54.33485412597656,
+      "eval_runtime": 0.8261,
+      "eval_samples_per_second": 55.684,
+      "eval_steps_per_second": 3.632,
+      "step": 21
+    },
+    {
+      "epoch": 6.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 70.2849349975586,
+      "eval_runtime": 0.8977,
+      "eval_samples_per_second": 51.24,
+      "eval_steps_per_second": 3.342,
+      "step": 24
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 57.703346252441406,
+      "eval_runtime": 0.7722,
+      "eval_samples_per_second": 59.569,
+      "eval_steps_per_second": 3.885,
+      "step": 28
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 0.375,
+      "loss": 64.7194,
+      "step": 30
+    },
+    {
+      "epoch": 8.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 115.52566528320312,
+      "eval_runtime": 0.7522,
+      "eval_samples_per_second": 61.153,
+      "eval_steps_per_second": 3.988,
+      "step": 31
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 72.79902648925781,
+      "eval_runtime": 0.9673,
+      "eval_samples_per_second": 47.556,
+      "eval_steps_per_second": 3.102,
+      "step": 35
+    },
+    {
+      "epoch": 10.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 41.867027282714844,
+      "eval_runtime": 0.7652,
+      "eval_samples_per_second": 60.112,
+      "eval_steps_per_second": 3.92,
+      "step": 38
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 0.3333333333333333,
+      "loss": 58.6249,
+      "step": 40
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 26.276500701904297,
+      "eval_runtime": 0.7682,
+      "eval_samples_per_second": 59.882,
+      "eval_steps_per_second": 3.905,
+      "step": 42
+    },
+    {
+      "epoch": 12.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 41.724510192871094,
+      "eval_runtime": 0.7512,
+      "eval_samples_per_second": 61.237,
+      "eval_steps_per_second": 3.994,
+      "step": 45
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 23.69620704650879,
+      "eval_runtime": 0.7782,
+      "eval_samples_per_second": 59.112,
+      "eval_steps_per_second": 3.855,
+      "step": 49
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 0.2916666666666667,
+      "loss": 49.7372,
+      "step": 50
+    },
+    {
+      "epoch": 14.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 13.426469802856445,
+      "eval_runtime": 0.7452,
+      "eval_samples_per_second": 61.729,
+      "eval_steps_per_second": 4.026,
+      "step": 52
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 7.040547847747803,
+      "eval_runtime": 0.7313,
+      "eval_samples_per_second": 62.904,
+      "eval_steps_per_second": 4.102,
+      "step": 56
+    },
+    {
+      "epoch": 16.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 5.077681064605713,
+      "eval_runtime": 0.7837,
+      "eval_samples_per_second": 58.698,
+      "eval_steps_per_second": 3.828,
+      "step": 59
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 0.25,
+      "loss": 11.7669,
+      "step": 60
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 13.569025039672852,
+      "eval_runtime": 0.74,
+      "eval_samples_per_second": 62.165,
+      "eval_steps_per_second": 4.054,
+      "step": 63
+    },
+    {
+      "epoch": 18.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 5.542473316192627,
+      "eval_runtime": 0.7968,
+      "eval_samples_per_second": 57.731,
+      "eval_steps_per_second": 3.765,
+      "step": 66
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.20833333333333334,
+      "loss": 13.3323,
+      "step": 70
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 6.449064254760742,
+      "eval_runtime": 0.8092,
+      "eval_samples_per_second": 56.849,
+      "eval_steps_per_second": 3.708,
+      "step": 70
+    },
+    {
+      "epoch": 20.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 7.306576251983643,
+      "eval_runtime": 0.8232,
+      "eval_samples_per_second": 55.882,
+      "eval_steps_per_second": 3.644,
+      "step": 73
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 10.843117713928223,
+      "eval_runtime": 0.7588,
+      "eval_samples_per_second": 60.621,
+      "eval_steps_per_second": 3.954,
+      "step": 77
+    },
+    {
+      "epoch": 22.86,
+      "learning_rate": 0.16666666666666666,
+      "loss": 9.2763,
+      "step": 80
+    },
+    {
+      "epoch": 22.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 12.158766746520996,
+      "eval_runtime": 0.7587,
+      "eval_samples_per_second": 60.63,
+      "eval_steps_per_second": 3.954,
+      "step": 80
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 5.4926228523254395,
+      "eval_runtime": 0.8528,
+      "eval_samples_per_second": 53.941,
+      "eval_steps_per_second": 3.518,
+      "step": 84
+    },
+    {
+      "epoch": 24.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 4.468862056732178,
+      "eval_runtime": 0.7743,
+      "eval_samples_per_second": 59.407,
+      "eval_steps_per_second": 3.874,
+      "step": 87
+    },
+    {
+      "epoch": 25.71,
+      "learning_rate": 0.125,
+      "loss": 6.8526,
+      "step": 90
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 3.7880265712738037,
+      "eval_runtime": 0.8747,
+      "eval_samples_per_second": 52.588,
+      "eval_steps_per_second": 3.43,
+      "step": 91
+    },
+    {
+      "epoch": 26.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.329732656478882,
+      "eval_runtime": 0.8285,
+      "eval_samples_per_second": 55.523,
+      "eval_steps_per_second": 3.621,
+      "step": 94
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 2.8531885147094727,
+      "eval_runtime": 0.744,
+      "eval_samples_per_second": 61.829,
+      "eval_steps_per_second": 4.032,
+      "step": 98
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 0.08333333333333333,
+      "loss": 3.0687,
+      "step": 100
+    },
+    {
+      "epoch": 28.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 2.694334030151367,
+      "eval_runtime": 0.8002,
+      "eval_samples_per_second": 57.483,
+      "eval_steps_per_second": 3.749,
+      "step": 101
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 2.0957062244415283,
+      "eval_runtime": 0.7804,
+      "eval_samples_per_second": 58.944,
+      "eval_steps_per_second": 3.844,
+      "step": 105
+    },
+    {
+      "epoch": 30.86,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.4001126289367676,
+      "eval_runtime": 0.853,
+      "eval_samples_per_second": 53.925,
+      "eval_steps_per_second": 3.517,
+      "step": 108
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 0.041666666666666664,
+      "loss": 2.1059,
+      "step": 110
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.308089017868042,
+      "eval_runtime": 0.8485,
+      "eval_samples_per_second": 54.211,
+      "eval_steps_per_second": 3.536,
+      "step": 112
+    },
+    {
+      "epoch": 32.86,
+      "eval_accuracy": 0.32608695652173914,
+      "eval_loss": 1.2391899824142456,
+      "eval_runtime": 0.8179,
+      "eval_samples_per_second": 56.243,
+      "eval_steps_per_second": 3.668,
+      "step": 115
+    },
+    {
+      "epoch": 34.0,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.2510451078414917,
+      "eval_runtime": 0.8127,
+      "eval_samples_per_second": 56.602,
+      "eval_steps_per_second": 3.691,
+      "step": 119
+    },
+    {
+      "epoch": 34.29,
+      "learning_rate": 0.0,
+      "loss": 1.3417,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "eval_accuracy": 0.45652173913043476,
+      "eval_loss": 1.233820915222168,
+      "eval_runtime": 0.787,
+      "eval_samples_per_second": 58.449,
+      "eval_steps_per_second": 3.812,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "step": 120,
+      "total_flos": 2.3770905934823424e+17,
+      "train_loss": 78.35383641719818,
+      "train_runtime": 207.9228,
+      "train_samples_per_second": 40.977,
+      "train_steps_per_second": 0.577
+    }
+  ],
+  "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
+}