Boya1_RMSProp_1-e5_20Epoch_swin-base-window7-224-in22k_fold3

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: 2.3110
• Accuracy: 0.6651

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
1.0919	1.0	923	1.1489	0.6081
0.9736	2.0	1846	1.0024	0.6559
0.975	3.0	2769	0.9707	0.6684
0.7334	4.0	3692	0.9560	0.6781
0.485	5.0	4615	1.0348	0.6662
0.316	6.0	5538	1.0693	0.6719
0.3512	7.0	6461	1.1640	0.6659
0.2952	8.0	7384	1.3227	0.6546
0.2423	9.0	8307	1.3837	0.6643
0.1889	10.0	9230	1.4884	0.6619
0.1312	11.0	10153	1.6309	0.6632
0.102	12.0	11076	1.7865	0.6622
0.0735	13.0	11999	1.8485	0.6586
0.1241	14.0	12922	2.0117	0.6600
0.1724	15.0	13845	2.0571	0.6627
0.0383	16.0	14768	2.1327	0.6603
0.0264	17.0	15691	2.2234	0.6641
0.0681	18.0	16614	2.2755	0.6681
0.0199	19.0	17537	2.3061	0.6643
0.1052	20.0	18460	2.3110	0.6651

Framework versions

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