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: 0.2978
• Accuracy: 0.9081

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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.4824	1.0	21	2.2558	0.2420
2.1213	2.0	42	1.8165	0.4170
1.6613	3.0	63	1.3678	0.5671
1.3237	4.0	84	1.1940	0.6060
1.1543	5.0	105	0.9205	0.7049
0.9317	6.0	126	0.8121	0.7314
0.7891	7.0	147	0.6553	0.7986
0.6812	8.0	168	0.5720	0.8180
0.6348	9.0	189	0.5364	0.8180
0.5488	10.0	210	0.4780	0.8428
0.505	11.0	231	0.4540	0.8569
0.4758	12.0	252	0.3992	0.8852
0.4306	13.0	273	0.4280	0.8675
0.3952	14.0	294	0.4019	0.8781
0.3726	15.0	315	0.3794	0.8763
0.3191	16.0	336	0.3482	0.8958
0.3014	17.0	357	0.3372	0.8940
0.2785	18.0	378	0.3472	0.8993
0.2948	19.0	399	0.3246	0.9064
0.2618	20.0	420	0.3060	0.9081
0.2705	21.0	441	0.3122	0.9046
0.2479	22.0	462	0.3061	0.9028
0.2411	23.0	483	0.3040	0.9099
0.2556	24.0	504	0.2990	0.9099
0.2413	25.0	525	0.2978	0.9081

Framework versions

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