cards-top_left_swin-tiny-patch4-window7-224-finetuned-v3_more_data

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: 0.9722
• Accuracy: 0.5941

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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.5501	1.0	1346	1.2633	0.4645
1.4882	2.0	2692	1.1866	0.4951
1.5148	3.0	4038	1.1644	0.5066
1.4605	4.0	5384	1.1546	0.5105
1.425	5.0	6730	1.0940	0.5361
1.4452	6.0	8076	1.0750	0.5530
1.4507	7.0	9422	1.0997	0.5301
1.4435	8.0	10768	1.0835	0.5445
1.3904	9.0	12114	1.0587	0.5493
1.3826	10.0	13460	1.0434	0.5581
1.4186	11.0	14806	1.0515	0.5536
1.3938	12.0	16152	1.0283	0.5635
1.3763	13.0	17498	1.0140	0.5740
1.3873	14.0	18844	1.0557	0.5470
1.3833	15.0	20190	1.0244	0.5638
1.385	16.0	21536	1.0345	0.5584
1.3492	17.0	22882	0.9997	0.5757
1.3332	18.0	24228	1.0106	0.5697
1.399	19.0	25574	0.9867	0.5846
1.3117	20.0	26920	0.9929	0.5833
1.362	21.0	28266	0.9895	0.5861
1.3279	22.0	29612	0.9853	0.5858
1.3057	23.0	30958	0.9872	0.5865
1.3217	24.0	32304	0.9761	0.5909
1.2854	25.0	33650	0.9800	0.5910
1.3194	26.0	34996	0.9867	0.5901
1.2733	27.0	36342	0.9927	0.5871
1.2949	28.0	37688	0.9755	0.5939
1.2836	29.0	39034	0.9738	0.5940
1.2974	30.0	40380	0.9722	0.5941

Framework versions

• Transformers 4.37.2
• Pytorch 2.0.1+cu117
• Datasets 2.17.0
• Tokenizers 0.15.2