End of training

README.md

@@ -22,7 +22,7 @@ model-index:
  metrics:
  - name: Accuracy
  type: accuracy
- value: 0.96
+ value: 0.9714285714285714
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1077
-- Accuracy: 0.96
+- Loss: 0.1138
+- Accuracy: 0.9714
 
 ## Model description
 

all_results.json

@@ -1,8 +1,13 @@
 {
- "epoch": 2.88,
- "total_flos": 1.1314446186361651e+17,
- "train_loss": 0.200619046886762,
- "train_runtime": 608.5465,
- "train_samples_per_second": 7.75,
- "train_steps_per_second": 0.059
+ "epoch": 9.6,
+ "eval_accuracy": 0.9714285714285714,
+ "eval_loss": 0.11381589621305466,
+ "eval_runtime": 1.3926,
+ "eval_samples_per_second": 125.668,
+ "eval_steps_per_second": 4.309,
+ "total_flos": 3.7552428159831245e+17,
+ "train_loss": 0.10836825122435888,
+ "train_runtime": 297.5248,
+ "train_samples_per_second": 52.836,
+ "train_steps_per_second": 0.403
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+ "epoch": 9.6,
+ "eval_accuracy": 0.9714285714285714,
+ "eval_loss": 0.11381589621305466,
+ "eval_runtime": 1.3926,
+ "eval_samples_per_second": 125.668,
+ "eval_steps_per_second": 4.309
+}

runs/Nov11_02-53-20_43d0f91f1623/events.out.tfevents.1699671618.43d0f91f1623.665.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddfcba77400905ddc5dc59aeba4e5d6fcb1bebf5a66d399b7d3e2d4d6d5328b3
+size 405

train_results.json

@@ -1,8 +1,8 @@
 {
- "epoch": 2.88,
- "total_flos": 1.1314446186361651e+17,
- "train_loss": 0.200619046886762,
- "train_runtime": 608.5465,
- "train_samples_per_second": 7.75,
- "train_steps_per_second": 0.059
+ "epoch": 9.6,
+ "total_flos": 3.7552428159831245e+17,
+ "train_loss": 0.10836825122435888,
+ "train_runtime": 297.5248,
+ "train_samples_per_second": 52.836,
+ "train_steps_per_second": 0.403
 }

trainer_state.json

@@ -1,73 +1,190 @@
 {
- "best_metric": 0.9657142857142857,
+ "best_metric": 0.9714285714285714,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-soccer-binary/checkpoint-12",
- "epoch": 2.88,
+ "epoch": 9.6,
  "eval_steps": 500,
- "global_step": 36,
+ "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
  {
  "epoch": 0.8,
- "learning_rate": 4.0625000000000005e-05,
- "loss": 0.3683,
+ "learning_rate": 4.166666666666667e-05,
+ "loss": 0.1286,
  "step": 10
  },
  {
  "epoch": 0.96,
- "eval_accuracy": 0.9657142857142857,
- "eval_loss": 0.12774832546710968,
- "eval_runtime": 52.6033,
- "eval_samples_per_second": 3.327,
- "eval_steps_per_second": 0.114,
+ "eval_accuracy": 0.9714285714285714,
+ "eval_loss": 0.11381589621305466,
+ "eval_runtime": 1.4093,
+ "eval_samples_per_second": 124.175,
+ "eval_steps_per_second": 4.257,
  "step": 12
  },
  {
  "epoch": 1.6,
- "learning_rate": 2.5e-05,
- "loss": 0.1466,
+ "learning_rate": 4.62962962962963e-05,
+ "loss": 0.1267,
  "step": 20
  },
  {
  "epoch": 2.0,
- "eval_accuracy": 0.96,
- "eval_loss": 0.13342900574207306,
- "eval_runtime": 1.4403,
- "eval_samples_per_second": 121.502,
- "eval_steps_per_second": 4.166,
+ "eval_accuracy": 0.9657142857142857,
+ "eval_loss": 0.12827394902706146,
+ "eval_runtime": 1.5512,
+ "eval_samples_per_second": 112.818,
+ "eval_steps_per_second": 3.868,
  "step": 25
  },
  {
  "epoch": 2.4,
- "learning_rate": 9.375000000000001e-06,
- "loss": 0.1313,
+ "learning_rate": 4.166666666666667e-05,
+ "loss": 0.121,
  "step": 30
  },
  {
- "epoch": 2.88,
+ "epoch": 2.96,
+ "eval_accuracy": 0.9657142857142857,
+ "eval_loss": 0.11240333318710327,
+ "eval_runtime": 1.3526,
+ "eval_samples_per_second": 129.378,
+ "eval_steps_per_second": 4.436,
+ "step": 37
+ },
+ {
+ "epoch": 3.2,
+ "learning_rate": 3.7037037037037037e-05,
+ "loss": 0.1174,
+ "step": 40
+ },
+ {
+ "epoch": 4.0,
+ "learning_rate": 3.240740740740741e-05,
+ "loss": 0.1142,
+ "step": 50
+ },
+ {
+ "epoch": 4.0,
+ "eval_accuracy": 0.9657142857142857,
+ "eval_loss": 0.11509174853563309,
+ "eval_runtime": 1.5727,
+ "eval_samples_per_second": 111.277,
+ "eval_steps_per_second": 3.815,
+ "step": 50
+ },
+ {
+ "epoch": 4.8,
+ "learning_rate": 2.777777777777778e-05,
+ "loss": 0.1069,
+ "step": 60
+ },
+ {
+ "epoch": 4.96,
+ "eval_accuracy": 0.96,
+ "eval_loss": 0.10633311420679092,
+ "eval_runtime": 1.3515,
+ "eval_samples_per_second": 129.483,
+ "eval_steps_per_second": 4.439,
+ "step": 62
+ },
+ {
+ "epoch": 5.6,
+ "learning_rate": 2.314814814814815e-05,
+ "loss": 0.1038,
+ "step": 70
+ },
+ {
+ "epoch": 6.0,
+ "eval_accuracy": 0.96,
+ "eval_loss": 0.12104564160108566,
+ "eval_runtime": 1.5722,
+ "eval_samples_per_second": 111.311,
+ "eval_steps_per_second": 3.816,
+ "step": 75
+ },
+ {
+ "epoch": 6.4,
+ "learning_rate": 1.8518518518518518e-05,
+ "loss": 0.0935,
+ "step": 80
+ },
+ {
+ "epoch": 6.96,
+ "eval_accuracy": 0.96,
+ "eval_loss": 0.11504363268613815,
+ "eval_runtime": 1.3905,
+ "eval_samples_per_second": 125.85,
+ "eval_steps_per_second": 4.315,
+ "step": 87
+ },
+ {
+ "epoch": 7.2,
+ "learning_rate": 1.388888888888889e-05,
+ "loss": 0.1006,
+ "step": 90
+ },
+ {
+ "epoch": 8.0,
+ "learning_rate": 9.259259259259259e-06,
+ "loss": 0.1042,
+ "step": 100
+ },
+ {
+ "epoch": 8.0,
  "eval_accuracy": 0.9657142857142857,
- "eval_loss": 0.11681389808654785,
- "eval_runtime": 1.7227,
- "eval_samples_per_second": 101.586,
- "eval_steps_per_second": 3.483,
- "step": 36
- },
- {
- "epoch": 2.88,
- "step": 36,
- "total_flos": 1.1314446186361651e+17,
- "train_loss": 0.200619046886762,
- "train_runtime": 608.5465,
- "train_samples_per_second": 7.75,
- "train_steps_per_second": 0.059
+ "eval_loss": 0.10376063734292984,
+ "eval_runtime": 1.7472,
+ "eval_samples_per_second": 100.161,
+ "eval_steps_per_second": 3.434,
+ "step": 100
+ },
+ {
+ "epoch": 8.8,
+ "learning_rate": 4.6296296296296296e-06,
+ "loss": 0.0945,
+ "step": 110
+ },
+ {
+ "epoch": 8.96,
+ "eval_accuracy": 0.96,
+ "eval_loss": 0.10713567584753036,
+ "eval_runtime": 1.5766,
+ "eval_samples_per_second": 110.995,
+ "eval_steps_per_second": 3.806,
+ "step": 112
+ },
+ {
+ "epoch": 9.6,
+ "learning_rate": 0.0,
+ "loss": 0.0891,
+ "step": 120
+ },
+ {
+ "epoch": 9.6,
+ "eval_accuracy": 0.96,
+ "eval_loss": 0.10769099742174149,
+ "eval_runtime": 1.4075,
+ "eval_samples_per_second": 124.337,
+ "eval_steps_per_second": 4.263,
+ "step": 120
+ },
+ {
+ "epoch": 9.6,
+ "step": 120,
+ "total_flos": 3.7552428159831245e+17,
+ "train_loss": 0.10836825122435888,
+ "train_runtime": 297.5248,
+ "train_samples_per_second": 52.836,
+ "train_steps_per_second": 0.403
  }
  ],
  "logging_steps": 10,
- "max_steps": 36,
- "num_train_epochs": 3,
+ "max_steps": 120,
+ "num_train_epochs": 10,
  "save_steps": 500,
- "total_flos": 1.1314446186361651e+17,
+ "total_flos": 3.7552428159831245e+17,
  "trial_name": null,
  "trial_params": null
 }