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: 7.8569
+- Loss: 7.9433
 - Accuracy: 0.1087
 
 ## Model description

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.10869565217391304,
+    "eval_loss": 7.943281173706055,
+    "eval_runtime": 0.8479,
+    "eval_samples_per_second": 54.252,
+    "eval_steps_per_second": 3.538,
+    "train_loss": 7.8947497049967446,
+    "train_runtime": 194.3282,
+    "train_samples_per_second": 43.843,
+    "train_steps_per_second": 0.618
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 34.29,
+    "eval_accuracy": 0.10869565217391304,
+    "eval_loss": 7.943281173706055,
+    "eval_runtime": 0.8479,
+    "eval_samples_per_second": 54.252,
+    "eval_steps_per_second": 3.538
+}

runs/Dec02_21-34-42_DESKTOP-SKBE9FB/events.out.tfevents.1733197099.DESKTOP-SKBE9FB.20780.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee5317e2bd6293d02ea48c3304d968d9b61592ff4be8fc888c95d3cab335f40a
+size 405

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 34.29,
+    "train_loss": 7.8947497049967446,
+    "train_runtime": 194.3282,
+    "train_samples_per_second": 43.843,
+    "train_steps_per_second": 0.618
+}

trainer_state.json

@@ -0,0 +1,417 @@
+{
+  "best_metric": 0.10869565217391304,
+  "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-15e-6\\checkpoint-3",
+  "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": 7.943281173706055,
+      "eval_runtime": 0.6567,
+      "eval_samples_per_second": 70.052,
+      "eval_steps_per_second": 4.569,
+      "step": 3
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.942506790161133,
+      "eval_runtime": 0.6662,
+      "eval_samples_per_second": 69.052,
+      "eval_steps_per_second": 4.503,
+      "step": 7
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 1.25e-06,
+      "loss": 7.9661,
+      "step": 10
+    },
+    {
+      "epoch": 2.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.94142484664917,
+      "eval_runtime": 0.6582,
+      "eval_samples_per_second": 69.892,
+      "eval_steps_per_second": 4.558,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.939181327819824,
+      "eval_runtime": 0.6802,
+      "eval_samples_per_second": 67.63,
+      "eval_steps_per_second": 4.411,
+      "step": 14
+    },
+    {
+      "epoch": 4.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.937195301055908,
+      "eval_runtime": 0.6577,
+      "eval_samples_per_second": 69.946,
+      "eval_steps_per_second": 4.562,
+      "step": 17
+    },
+    {
+      "epoch": 5.71,
+      "learning_rate": 1.388888888888889e-06,
+      "loss": 7.9068,
+      "step": 20
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.934326648712158,
+      "eval_runtime": 0.6435,
+      "eval_samples_per_second": 71.488,
+      "eval_steps_per_second": 4.662,
+      "step": 21
+    },
+    {
+      "epoch": 6.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.932020664215088,
+      "eval_runtime": 0.6717,
+      "eval_samples_per_second": 68.487,
+      "eval_steps_per_second": 4.467,
+      "step": 24
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.928760528564453,
+      "eval_runtime": 0.6447,
+      "eval_samples_per_second": 71.356,
+      "eval_steps_per_second": 4.654,
+      "step": 28
+    },
+    {
+      "epoch": 8.57,
+      "learning_rate": 1.25e-06,
+      "loss": 7.7849,
+      "step": 30
+    },
+    {
+      "epoch": 8.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.926177978515625,
+      "eval_runtime": 0.6401,
+      "eval_samples_per_second": 71.859,
+      "eval_steps_per_second": 4.686,
+      "step": 31
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.922581672668457,
+      "eval_runtime": 0.6371,
+      "eval_samples_per_second": 72.197,
+      "eval_steps_per_second": 4.709,
+      "step": 35
+    },
+    {
+      "epoch": 10.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.919796466827393,
+      "eval_runtime": 0.6246,
+      "eval_samples_per_second": 73.641,
+      "eval_steps_per_second": 4.803,
+      "step": 38
+    },
+    {
+      "epoch": 11.43,
+      "learning_rate": 1.111111111111111e-06,
+      "loss": 8.0613,
+      "step": 40
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.915912628173828,
+      "eval_runtime": 0.6592,
+      "eval_samples_per_second": 69.785,
+      "eval_steps_per_second": 4.551,
+      "step": 42
+    },
+    {
+      "epoch": 12.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.912907600402832,
+      "eval_runtime": 0.6392,
+      "eval_samples_per_second": 71.97,
+      "eval_steps_per_second": 4.694,
+      "step": 45
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.908822059631348,
+      "eval_runtime": 0.6707,
+      "eval_samples_per_second": 68.589,
+      "eval_steps_per_second": 4.473,
+      "step": 49
+    },
+    {
+      "epoch": 14.29,
+      "learning_rate": 9.722222222222222e-07,
+      "loss": 7.827,
+      "step": 50
+    },
+    {
+      "epoch": 14.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.905688762664795,
+      "eval_runtime": 0.6517,
+      "eval_samples_per_second": 70.589,
+      "eval_steps_per_second": 4.604,
+      "step": 52
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.901515960693359,
+      "eval_runtime": 0.6767,
+      "eval_samples_per_second": 67.981,
+      "eval_steps_per_second": 4.434,
+      "step": 56
+    },
+    {
+      "epoch": 16.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.898391246795654,
+      "eval_runtime": 0.6386,
+      "eval_samples_per_second": 72.027,
+      "eval_steps_per_second": 4.697,
+      "step": 59
+    },
+    {
+      "epoch": 17.14,
+      "learning_rate": 8.333333333333334e-07,
+      "loss": 7.8359,
+      "step": 60
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.894200801849365,
+      "eval_runtime": 0.6391,
+      "eval_samples_per_second": 71.971,
+      "eval_steps_per_second": 4.694,
+      "step": 63
+    },
+    {
+      "epoch": 18.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.891049385070801,
+      "eval_runtime": 0.6557,
+      "eval_samples_per_second": 70.159,
+      "eval_steps_per_second": 4.576,
+      "step": 66
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 6.944444444444445e-07,
+      "loss": 7.9032,
+      "step": 70
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.886945724487305,
+      "eval_runtime": 0.6667,
+      "eval_samples_per_second": 69.001,
+      "eval_steps_per_second": 4.5,
+      "step": 70
+    },
+    {
+      "epoch": 20.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.883902072906494,
+      "eval_runtime": 0.6817,
+      "eval_samples_per_second": 67.482,
+      "eval_steps_per_second": 4.401,
+      "step": 73
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.879933834075928,
+      "eval_runtime": 0.6561,
+      "eval_samples_per_second": 70.106,
+      "eval_steps_per_second": 4.572,
+      "step": 77
+    },
+    {
+      "epoch": 22.86,
+      "learning_rate": 5.555555555555555e-07,
+      "loss": 7.9294,
+      "step": 80
+    },
+    {
+      "epoch": 22.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.8770551681518555,
+      "eval_runtime": 0.6772,
+      "eval_samples_per_second": 67.932,
+      "eval_steps_per_second": 4.43,
+      "step": 80
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.873494625091553,
+      "eval_runtime": 0.6541,
+      "eval_samples_per_second": 70.321,
+      "eval_steps_per_second": 4.586,
+      "step": 84
+    },
+    {
+      "epoch": 24.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.870996952056885,
+      "eval_runtime": 0.7142,
+      "eval_samples_per_second": 64.411,
+      "eval_steps_per_second": 4.201,
+      "step": 87
+    },
+    {
+      "epoch": 25.71,
+      "learning_rate": 4.166666666666667e-07,
+      "loss": 7.9185,
+      "step": 90
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.867926597595215,
+      "eval_runtime": 0.6451,
+      "eval_samples_per_second": 71.302,
+      "eval_steps_per_second": 4.65,
+      "step": 91
+    },
+    {
+      "epoch": 26.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.865866184234619,
+      "eval_runtime": 0.7027,
+      "eval_samples_per_second": 65.465,
+      "eval_steps_per_second": 4.269,
+      "step": 94
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.863403797149658,
+      "eval_runtime": 0.6471,
+      "eval_samples_per_second": 71.081,
+      "eval_steps_per_second": 4.636,
+      "step": 98
+    },
+    {
+      "epoch": 28.57,
+      "learning_rate": 2.7777777777777776e-07,
+      "loss": 7.7948,
+      "step": 100
+    },
+    {
+      "epoch": 28.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.861809730529785,
+      "eval_runtime": 0.6818,
+      "eval_samples_per_second": 67.467,
+      "eval_steps_per_second": 4.4,
+      "step": 101
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.860015869140625,
+      "eval_runtime": 0.656,
+      "eval_samples_per_second": 70.125,
+      "eval_steps_per_second": 4.573,
+      "step": 105
+    },
+    {
+      "epoch": 30.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.858927249908447,
+      "eval_runtime": 0.6892,
+      "eval_samples_per_second": 66.747,
+      "eval_steps_per_second": 4.353,
+      "step": 108
+    },
+    {
+      "epoch": 31.43,
+      "learning_rate": 1.3888888888888888e-07,
+      "loss": 7.906,
+      "step": 110
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.857845783233643,
+      "eval_runtime": 0.6391,
+      "eval_samples_per_second": 71.971,
+      "eval_steps_per_second": 4.694,
+      "step": 112
+    },
+    {
+      "epoch": 32.86,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.857306003570557,
+      "eval_runtime": 0.6662,
+      "eval_samples_per_second": 69.052,
+      "eval_steps_per_second": 4.503,
+      "step": 115
+    },
+    {
+      "epoch": 34.0,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.856945514678955,
+      "eval_runtime": 0.6612,
+      "eval_samples_per_second": 69.575,
+      "eval_steps_per_second": 4.538,
+      "step": 119
+    },
+    {
+      "epoch": 34.29,
+      "learning_rate": 0.0,
+      "loss": 7.9031,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "eval_accuracy": 0.10869565217391304,
+      "eval_loss": 7.856921195983887,
+      "eval_runtime": 0.6551,
+      "eval_samples_per_second": 70.213,
+      "eval_steps_per_second": 4.579,
+      "step": 120
+    },
+    {
+      "epoch": 34.29,
+      "step": 120,
+      "total_flos": 2.3770905934823424e+17,
+      "train_loss": 7.8947497049967446,
+      "train_runtime": 194.3282,
+      "train_samples_per_second": 43.843,
+      "train_steps_per_second": 0.618
+    }
+  ],
+  "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
+}