Modelpublic

system-admin/swin-tiny-patch4-window7-224-finetuned-student_two_classes

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes6downloads

Model Card

swin-tiny-patch4-window7-224-finetuned-studenttwoclasses

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: 0.6649
—Accuracy: 0.8

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-06
—trainbatchsize: 32
—evalbatchsize: 32
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3416	1.0	13	0.6767	0.81
0.2745	2.0	26	0.7166	0.81
0.2482	3.0	39	0.7644	0.83
0.2664	4.0	52	0.7142	0.81
0.2372	5.0	65	0.8010	0.85
0.2238	6.0	78	0.8655	0.86
0.1687	7.0	91	0.8156	0.83
0.209	8.0	104	0.8460	0.83
0.4251	9.0	117	0.7301	0.81
0.4392	10.0	130	0.6775	0.78
0.3285	11.0	143	0.7145	0.81
0.3178	12.0	156	0.7431	0.83
0.4715	13.0	169	0.6973	0.81
0.373	14.0	182	0.6912	0.81
0.3378	15.0	195	0.7018	0.82
0.3867	16.0	208	0.6885	0.81
0.3525	17.0	221	0.6761	0.81
0.4253	18.0	234	0.6643	0.8
0.325	19.0	247	0.6643	0.8
0.372	20.0	260	0.6649	0.8

Framework versions

—Transformers 4.40.1
—Pytorch 2.2.1+cu121
—Datasets 2.19.0
—Tokenizers 0.19.1