cards-top_right_swin-tiny-patch4-window7-224-finetuned-v2_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: 0.9268
• Accuracy: 0.6269

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
1.4585	1.0	1363	1.2999	0.4337
1.4211	2.0	2726	1.1663	0.4927
1.4203	3.0	4089	1.0770	0.5312
1.4669	4.0	5453	1.0744	0.5496
1.3781	5.0	6816	1.0245	0.5599
1.3852	6.0	8179	1.0645	0.5402
1.3407	7.0	9542	1.0011	0.5696
1.3727	8.0	10906	0.9898	0.5801
1.328	9.0	12269	0.9965	0.5738
1.3374	10.0	13632	0.9722	0.5874
1.3513	11.0	14995	0.9632	0.5873
1.3728	12.0	16359	0.9818	0.5802
1.3289	13.0	17722	0.9845	0.5729
1.3219	14.0	19085	0.9633	0.5881
1.2893	15.0	20448	0.9312	0.6004
1.3088	16.0	21812	0.9537	0.5903
1.3252	17.0	23175	0.9432	0.5986
1.3424	18.0	24538	0.9291	0.5979
1.3077	19.0	25901	0.9245	0.6020
1.2466	20.0	27265	0.9304	0.6039
1.2767	21.0	28628	0.9122	0.6099
1.2553	22.0	29991	0.9312	0.6005
1.2698	23.0	31354	0.9137	0.6092
1.2591	24.0	32718	0.9113	0.6134
1.277	25.0	34081	0.9095	0.6142
1.2742	26.0	35444	0.9227	0.6100
1.222	27.0	36807	0.9090	0.6147
1.2368	28.0	38171	0.9020	0.6172
1.198	29.0	39534	0.9071	0.6157
1.2076	30.0	40897	0.9031	0.6214
1.212	31.0	42260	0.9136	0.6175
1.2105	32.0	43624	0.9170	0.6151
1.2687	33.0	44987	0.9047	0.6186
1.2038	34.0	46350	0.9061	0.6190
1.1957	35.0	47713	0.9052	0.6255
1.1962	36.0	49077	0.9057	0.6210
1.1866	37.0	50440	0.9105	0.6227
1.2545	38.0	51803	0.9173	0.6206
1.1642	39.0	53166	0.9120	0.6239
1.1711	40.0	54530	0.9235	0.6177
1.2339	41.0	55893	0.9295	0.6143
1.1132	42.0	57256	0.9143	0.6234
1.1977	43.0	58619	0.9163	0.6256
1.1617	44.0	59983	0.9246	0.6233
1.1357	45.0	61346	0.9196	0.6255
1.1362	46.0	62709	0.9221	0.6259
1.1472	47.0	64072	0.9206	0.6263
1.184	48.0	65436	0.9282	0.6256
1.1096	49.0	66799	0.9252	0.6269
1.1425	49.99	68150	0.9268	0.6269

Framework versions

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