medsiglip-448-scin-classification

This model is a fine-tuned version of google/medsiglip-448 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.0076
• Roc Auc: 0.8266

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 5
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Roc Auc
9.4303	0.4961	40	1.0166	0.7371
7.8956	0.9922	80	0.9521	0.7859
6.3407	1.4837	120	0.9112	0.8071
5.9863	1.9798	160	0.9219	0.8193
4.2234	2.4713	200	1.0053	0.8266
3.9940	2.9674	240	1.0076	0.8266

Framework versions

• Transformers 5.0.0
• Pytorch 2.9.0+cu126
• Datasets 4.5.0
• Tokenizers 0.22.2