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: swin-tiny-patch4-window7-224-finetuned-isic217
    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: 0.5909090909090909

swin-tiny-patch4-window7-224-finetuned-isic217

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: 2.3724
• Accuracy: 0.5909

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• 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
2.2679	0.9796	24	2.1550	0.0909
2.0504	2.0	49	2.0559	0.2727
1.8943	2.9796	73	2.0186	0.2273
1.5671	4.0	98	1.8154	0.2273
1.3425	4.9796	122	2.0475	0.2273
1.2758	6.0	147	2.1914	0.2273
0.9808	6.9796	171	2.0478	0.3636
0.7246	8.0	196	1.8840	0.4091
0.7323	8.9796	220	2.1831	0.4091
0.4881	10.0	245	2.2868	0.3636
0.4346	10.9796	269	2.2312	0.4545
0.5647	12.0	294	1.9897	0.4091
0.1464	12.9796	318	2.0579	0.4545
0.5575	14.0	343	2.1859	0.4545
0.3894	14.9796	367	2.7353	0.3636
0.4326	16.0	392	2.4455	0.3636
0.3715	16.9796	416	2.3104	0.5455
0.3966	18.0	441	2.4597	0.4545
0.1855	18.9796	465	2.3335	0.3636
0.1528	20.0	490	2.3630	0.4091
0.2036	20.9796	514	2.3520	0.4545
0.2026	22.0	539	2.7012	0.4091
0.2127	22.9796	563	2.3724	0.5909
0.2719	24.0	588	3.0376	0.3182
0.1292	24.9796	612	2.5684	0.5
0.2533	26.0	637	2.6974	0.4091
0.1947	26.9796	661	2.6957	0.4091
0.1805	28.0	686	2.8953	0.4091
0.1123	28.9796	710	2.8240	0.4091
0.2143	30.0	735	2.3880	0.4545
0.1845	30.9796	759	2.6072	0.3636
0.0921	32.0	784	2.7256	0.4545
0.0276	32.9796	808	2.4074	0.4091
0.0876	34.0	833	2.6043	0.4545
0.0253	34.9796	857	2.7620	0.4545
0.1904	36.0	882	2.6911	0.4091
0.072	36.9796	906	2.6528	0.4545
0.169	38.0	931	2.6454	0.4545
0.0978	38.9796	955	2.6269	0.5
0.069	40.0	980	2.4154	0.4545
0.0159	40.9796	1004	2.7026	0.4545
0.2046	42.0	1029	2.5213	0.4545
0.0329	42.9796	1053	2.6399	0.5
0.0166	44.0	1078	2.7787	0.4545
0.0812	44.9796	1102	2.8176	0.4545
0.0197	46.0	1127	2.8049	0.4545
0.0989	46.9796	1151	2.7479	0.4545
0.054	48.0	1176	2.7614	0.4545
0.1095	48.9796	1200	2.7604	0.5

Framework versions

• Transformers 4.42.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1