Medsiglip 448 Ft Crc100k

--- library_name: transformers license: other base_model: google/medsiglip-448 tags: - generated_from_trainer datasets: - imagefolder model-index: - name: medsiglip-448-ft-crc100k results: [] --- <!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. --> # medsiglip-448-ft-crc100k This model is a fine-tuned version of [google/medsiglip-448](https://huggingface.co/google/medsiglip-448) on the imagefolder dataset. It achieves the following results on the evaluation set: - Loss: 1.3559 ## 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: 32 - eval_batch_size: 32 - seed: 42 - gradient_accumulation_steps: 8 - total_train_batch_size: 256 - 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: 2 ### Training results | Training Loss | Epoch | Step | Validation Loss | |:-------------:|:------:|:----:|:---------------:| | 2.2380 | 1.3972 | 50 | 1.3793 | | 2.2380 | 2.0 | 72 | 1.3559 | ### Framework versions - Transformers 5.12.1 - Pytorch 2.11.0+cu128 - Datasets 5.0.0 - Tokenizers 0.22.2

Source context: 4 downloads · 0 likes · Pipeline zero-shot-image-classification · Library transformers · Repo billzhonggz/medsiglip-448-ft-crc100k