swin-brain-abnormalities-classification-fold2

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: 0.0893
• Accuracy: 0.9688

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9549	0.9714	17	0.7206	0.7236
0.5426	2.0	35	0.3207	0.8686
0.3933	2.9714	52	0.2391	0.8970
0.2881	4.0	70	0.2387	0.8943
0.2524	4.9714	87	0.1485	0.9444
0.1997	6.0	105	0.1185	0.9566
0.1746	6.9714	122	0.1188	0.9553
0.1589	8.0	140	0.0990	0.9621
0.1369	8.9714	157	0.1044	0.9593
0.1281	10.0	175	0.0957	0.9675
0.1211	10.9714	192	0.1026	0.9607
0.1233	12.0	210	0.1258	0.9499
0.1141	12.9714	227	0.0861	0.9729
0.1021	14.0	245	0.0945	0.9648
0.1102	14.5714	255	0.0893	0.9688

Framework versions

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