|
--- |
|
datasets: |
|
- romprr/CXR_BioXAi_Hackathon_2024 |
|
language: |
|
- en |
|
base_model: |
|
- mistral-community/pixtral-12b |
|
--- |
|
|
|
### Model Description |
|
|
|
CXR-Pixtral, is a fine-tuned version of the Pixtral-12B (Mistral AI) model, specifically designed for Medical Report Generation (MRG) from chest X-ray images. It excels in generating detailed and accurate medical reports based on chest X-rays. |
|
|
|
Developed by: William Peoc'h (@williampeoch), Romain Poirrier (@romprr), Antoine Gorceix (@aggorceix), Nicolas GOSSARD and Mathieu Duteil |
|
|
|
Fine-tuned on the following dataset: romprr/CXR_BioXAi_Hackathon_2024 |
|
|
|
- **Finetuned from model :** [mistral-community/pixtral-12b] |
|
|
|
|
|
## How to Get Started with the Model |
|
|
|
Soon |
|
|
|
|
|
## Training Details |
|
|
|
### Training Data |
|
|
|
Fine-tuned on the following dataset: [romprr/CXR_BioXAi_Hackathon_202] |
