Hugging Face's logo

Ozantsk
/
biomedclip-rocov2-finetuned

OpenCLIP
biomedclip
rocov2
medical-image-retrieval
Model card Files
xet
 Community
How to use from the
Use from the
OpenCLIP library
import open_clip

model, preprocess_train, preprocess_val = open_clip.create_model_and_transforms('hf-hub:Ozantsk/biomedclip-rocov2-finetuned')
tokenizer = open_clip.get_tokenizer('hf-hub:Ozantsk/biomedclip-rocov2-finetuned')

BiomedCLIP fine-tuned on ROCOv2

Fine-tuned from microsoft/BiomedCLIP-PubMedBERT_256-vit_base_patch16_224 on eltorio/ROCOv2-radiology with CLIP contrastive image-caption loss.

Main hyperparameters:

  • LR: 1e-05
  • Epochs: 4
  • Batch size: 16
  • Gradient accumulation: 4
  • Effective batch size: 64
  • Context length: 256

Saved files:

  • open_clip_pytorch_model.bin
  • open_clip_config.json
  • training_config.json
Downloads last month
5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Ozantsk/biomedclip-rocov2-finetuned

Base model

microsoft/BiomedCLIP-PubMedBERT_256-vit_base_patch16_224
Finetuned
(7)
this model

Dataset used to train Ozantsk/biomedclip-rocov2-finetuned