Modelpublic

selmamalak/organsmnist-swin-base-finetuned

sourceHugging Faceapache-2.0updated 2y agoView on Hugging Face

0likes10downloads

Model Card

organsmnist-swin-base-finetuned

This model is a fine-tuned version of microsoft/swin-large-patch4-window7-224-in22k on the medmnist-v2 dataset. It achieves the following results on the evaluation set:

—Loss: 0.4655
—Accuracy: 0.8230
—Precision: 0.7898
—Recall: 0.7786
—F1: 0.7831

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: 0.005
—trainbatchsize: 16
—evalbatchsize: 16
—seed: 42
—gradientaccumulationsteps: 4
—totaltrainbatch_size: 64
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 10
—mixedprecisiontraining: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.918	1.0	218	0.4249	0.8356	0.8139	0.7795	0.7529
0.889	2.0	436	0.4593	0.8263	0.7916	0.7695	0.7668
0.8566	3.0	654	0.3293	0.8748	0.8505	0.8205	0.8088
0.7781	4.0	872	0.3455	0.8679	0.8313	0.8088	0.7921
0.7241	5.0	1090	0.3565	0.8691	0.8758	0.8110	0.7902
0.6568	6.0	1308	0.3337	0.8809	0.8458	0.8295	0.8081
0.5643	7.0	1526	0.2581	0.8972	0.8392	0.8377	0.8332
0.5834	8.0	1744	0.2706	0.8956	0.8402	0.8360	0.8319
0.4771	9.0	1962	0.2721	0.9001	0.8524	0.8445	0.8364
0.5102	10.0	2180	0.2898	0.9009	0.8460	0.8410	0.8407

Framework versions

—PEFT 0.11.1
—Transformers 4.39.3
—Pytorch 2.1.2
—Datasets 2.18.0
—Tokenizers 0.15.2