Medical Generation Model

Overview

This repository contains a fine-tuned T5 model designed to generate medical diagnoses and treatment recommendations. The model was trained on clinical scenarios to provide accurate and contextually relevant medical outputs based on input prompts.

Model Details

  • Model Type: T5
  • Tokenizer: T5 tokenizer
  • Training Data: Clinical scenarios and medical texts

Installation

To use this model, install the required libraries with pip:

pip install transformers
pip install tensorflow

# Load the fine-tuned model and tokenizer
from transformers import T5Tokenizer, TFT5ForConditionalGeneration

model_id = "Ra-Is/medical-gen-small"
model = TFT5ForConditionalGeneration.from_pretrained(model_id)
tokenizer = T5Tokenizer.from_pretrained(model_id)

# Prepare a sample input prompt
input_prompt = ("A 35-year-old female presents with a 2-week history of "
                "persistent cough, shortness of breath, and fatigue. She has "
                "a history of asthma and has recently been exposed to a sick "
                "family member with a respiratory infection. Chest X-ray shows "
                "bilateral infiltrates. What is the likely diagnosis, and what "
                "should be the treatment?")

# Tokenize the input
input_ids = tokenizer(input_prompt, return_tensors="tf").input_ids

# Generate the output (diagnosis)
outputs = model.generate(
        input_ids,
        max_length=512,
        num_beams=5,
        temperature=1,
        top_k=50,
        top_p=0.9,
        do_sample=True,  # Enable sampling
        early_stopping=True
    )

# Decode and print the output
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(generated_text)
Downloads last month
9
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Ra-Is/medical-gen-small

Base model

google-t5/t5-small
Finetuned
(1628)
this model