swin-brain-tumor-type-classification

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

• Loss: 0.2385
• Accuracy: 0.9328

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-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 35

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.6739	1.0	19	2.4239	0.1978
2.2793	2.0	38	2.1011	0.3619
1.9273	3.0	57	1.6876	0.4963
1.5549	4.0	76	1.3626	0.5709
1.2386	5.0	95	1.1268	0.6269
1.1004	6.0	114	0.9445	0.7015
0.9008	7.0	133	0.8701	0.7313
0.8023	8.0	152	0.7917	0.7425
0.6566	9.0	171	0.6990	0.7612
0.6691	10.0	190	0.6204	0.7985
0.5605	11.0	209	0.5511	0.8321
0.5472	12.0	228	0.4945	0.8358
0.5098	13.0	247	0.4302	0.8619
0.4362	14.0	266	0.4027	0.8843
0.416	15.0	285	0.3956	0.8657
0.4095	16.0	304	0.3605	0.8881
0.3577	17.0	323	0.3339	0.8918
0.3624	18.0	342	0.3883	0.8694
0.304	19.0	361	0.3496	0.8769
0.2784	20.0	380	0.3275	0.8806
0.2763	21.0	399	0.3721	0.8806
0.2824	22.0	418	0.3156	0.8955
0.2453	23.0	437	0.3155	0.8843
0.2438	24.0	456	0.2928	0.9030
0.2285	25.0	475	0.2667	0.9216
0.2478	26.0	494	0.2816	0.9142
0.2242	27.0	513	0.2768	0.8993
0.2	28.0	532	0.2815	0.9142
0.2076	29.0	551	0.2443	0.9216
0.1978	30.0	570	0.2381	0.9216
0.1821	31.0	589	0.2563	0.9216
0.1786	32.0	608	0.2449	0.9254
0.1809	33.0	627	0.2385	0.9328
0.1812	34.0	646	0.2448	0.9291
0.1688	35.0	665	0.2446	0.9291

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0