jianna4/finetuned_disease

Fine-Tuned Falcon-7B for Medical Text Generation

This is a fine-tuned version of the Falcon-7B-Instruct model, adapted for generating medical text related to common diseases. The model has been fine-tuned using LoRA (Low-Rank Adaptation) on a dataset of medical texts.

Model Details

• Base Model: tiiuae/falcon-7b-instruct
• Fine-Tuning Method: LoRA (Low-Rank Adaptation)
• Quantization: 4-bit (using bitsandbytes)
• Training Dataset: Medical text data (common diseases)
• Training Framework: PyTorch with Hugging Face Transformers
• Fine-Tuning Duration: 3 epochs
• Learning Rate: 1e-3
• Batch Size: 2 (per device)

Usage

You can use this model for generating medical text or answering questions related to common diseases.

Using the Hugging Face Inference API

1. Install the transformers library:

   pip install transformers
   """