medsiglip-abdominal-pain-with-custom-head

This model is a fine-tuned version of google/medsiglip-448 on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 4.1069
• Accuracy: 0.479
• F1: 0.4896

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: 0.0001
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 64
• optimizer: Use adamw_torch 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: 50
• training_steps: 250

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.6127	0.2185	50	2.7326	0.453	0.4378
0.5664	0.4369	100	1.9900	0.533	0.5798
0.093	0.6554	150	2.2348	0.646	0.6535
0.0003	0.8738	200	4.1069	0.479	0.4896

Framework versions

• Transformers 4.54.1
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4