README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: swin-tiny-patch4-window7-224-shortSleeveCleanedData
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.994535519125683

swin-tiny-patch4-window7-224-shortSleeveCleanedData

This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.0355
• Accuracy: 0.9945

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 7
• total_train_batch_size: 56
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.01
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.1819	1.0	147	0.0471	0.9880
0.1431	2.0	294	0.0457	0.9891
0.1001	3.0	441	0.0392	0.9891
0.116	4.0	588	0.0451	0.9880
0.1144	5.0	735	0.0398	0.9902
0.0787	6.0	882	0.0441	0.9902
0.0998	7.0	1029	0.0320	0.9902
0.124	8.0	1176	0.0364	0.9902
0.103	9.0	1323	0.0395	0.9880
0.0591	10.0	1470	0.0299	0.9913
0.0445	11.0	1617	0.0302	0.9913
0.0684	12.0	1764	0.0350	0.9880
0.0358	13.0	1911	0.0408	0.9891
0.0548	14.0	2058	0.0382	0.9902
0.0611	15.0	2205	0.0331	0.9923
0.0231	16.0	2352	0.0355	0.9945
0.046	17.0	2499	0.0321	0.9934
0.0648	18.0	2646	0.0327	0.9923
0.0565	19.0	2793	0.0320	0.9923
0.0413	20.0	2940	0.0327	0.9923

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3