histoSwin-base

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the nctcrche100_k dataset. It achieves the following results on the evaluation set:

• Loss: 0.1260
• Accuracy: 0.9709

Model description

More information needed

Intended uses & limitations

For TMA classification.

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: 2
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0386	1.0	1562	0.1740	0.9592
0.05	2.0	3125	0.2313	0.9524
0.0324	3.0	4687	0.2606	0.9604
0.005	4.0	6250	0.1724	0.9660
0.0037	5.0	7810	0.1260	0.9709

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1
• Datasets 2.14.5
• Tokenizers 0.13.3