Boya1_RMSProp_1-e5_20Epoch_swin-base-window7-224-in22k_fold4

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224-in22k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 2.4944
• Accuracy: 0.6638

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.1419	1.0	924	1.1069	0.6193
0.8711	2.0	1848	1.0127	0.6435
0.7373	3.0	2772	0.9976	0.6565
0.8211	4.0	3696	0.9949	0.6684
0.6291	5.0	4620	1.0468	0.6735
0.3396	6.0	5544	1.1204	0.6646
0.3275	7.0	6468	1.2442	0.6586
0.3288	8.0	7392	1.3222	0.6594
0.2359	9.0	8316	1.4540	0.6657
0.2071	10.0	9240	1.5984	0.6581
0.112	11.0	10164	1.6998	0.6600
0.1118	12.0	11088	1.8535	0.6600
0.0722	13.0	12012	2.0369	0.6627
0.062	14.0	12936	2.1305	0.6567
0.0657	15.0	13860	2.2604	0.6616
0.0351	16.0	14784	2.3298	0.6608
0.0555	17.0	15708	2.4139	0.6613
0.0491	18.0	16632	2.4530	0.6638
0.0881	19.0	17556	2.4844	0.6646
0.0213	20.0	18480	2.4944	0.6638

Framework versions

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