README.md

metadata

license: apache-2.0
base_model: microsoft/swin-tiny-patch4-window7-224
tags:
  - generated_from_trainer
datasets:
  - imagefolder
metrics:
  - accuracy
model-index:
  - name: results
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: imagefolder
          type: imagefolder
          config: default
          split: validation
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6461538461538462

results

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: 0.8263
• Accuracy: 0.6462

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	1	1.1611	0.5077
No log	2.0	3	1.1465	0.5077
No log	3.0	5	1.1057	0.5077
No log	4.0	6	1.1134	0.5077
No log	5.0	7	1.0864	0.5077
No log	6.0	9	0.9838	0.5385
0.5404	7.0	11	0.9655	0.5538
0.5404	8.0	12	0.9630	0.5692
0.5404	9.0	13	0.9631	0.5538
0.5404	10.0	15	1.0177	0.5385
0.5404	11.0	17	1.0124	0.5538
0.5404	12.0	18	0.9905	0.5692
0.5404	13.0	19	0.9473	0.6154
0.5207	14.0	21	0.9549	0.6
0.5207	15.0	23	0.9348	0.5846
0.5207	16.0	24	0.9019	0.5846
0.5207	17.0	25	0.8687	0.5846
0.5207	18.0	27	0.8462	0.5846
0.5207	19.0	29	0.8418	0.6154
0.5146	20.0	30	0.8419	0.6
0.5146	21.0	31	0.8435	0.5692
0.5146	22.0	33	0.8415	0.5538
0.5146	23.0	35	0.8293	0.6154
0.5146	24.0	36	0.8254	0.6
0.5146	25.0	37	0.8219	0.6154
0.5146	26.0	39	0.8195	0.6462
0.4352	27.0	41	0.8192	0.6462
0.4352	28.0	42	0.8198	0.6308
0.4352	29.0	43	0.8230	0.6615
0.4352	30.0	45	0.8264	0.6462
0.4352	31.0	47	0.8268	0.6462
0.4352	32.0	48	0.8266	0.6462
0.4352	33.0	49	0.8263	0.6462
0.4724	33.33	50	0.8263	0.6462

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1