2024_08_16_swinv2-base-patch4-window8-256

This model was trained from scratch on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.4529
• Accuracy: 0.8068

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: 1e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 4
• 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
0.6942	1.0	87	0.6796	0.6818
0.6661	2.0	174	0.5973	0.7045
0.5666	3.0	261	0.5844	0.7045
0.6134	4.0	348	0.5798	0.7045
0.5207	5.0	435	0.5817	0.6932
0.75	6.0	522	0.5488	0.7159
0.4155	7.0	609	0.5373	0.7045
0.5122	8.0	696	0.5057	0.7386
0.722	9.0	783	0.4951	0.7045
0.5301	10.0	870	0.5349	0.7727
0.5881	11.0	957	0.4795	0.7955
0.5295	12.0	1044	0.4843	0.7955
0.6252	13.0	1131	0.4529	0.8068
0.9347	14.0	1218	0.4670	0.7955
0.5375	15.0	1305	0.4468	0.8068
0.4811	16.0	1392	0.4912	0.7841
0.5728	17.0	1479	0.4636	0.8068
0.7997	18.0	1566	0.4631	0.8068
0.4473	19.0	1653	0.4785	0.8068
0.4999	20.0	1740	0.5162	0.8068
0.4572	21.0	1827	0.5742	0.7955
0.2571	22.0	1914	0.5181	0.7955
0.5085	23.0	2001	0.4937	0.7955
0.7698	24.0	2088	0.4764	0.7955
0.558	25.0	2175	0.4742	0.8068
0.5462	26.0	2262	0.5320	0.7841
0.5218	27.0	2349	0.5298	0.7841
0.5228	28.0	2436	0.5182	0.7955
0.5787	29.0	2523	0.5104	0.8068
0.7511	30.0	2610	0.5152	0.8068

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.1+cu118
• Datasets 2.20.0
• Tokenizers 0.19.1