app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

# Load model (CPU-friendly, no token required)
model_id = "replit/replit-code-v1_5-3b"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# Ensure it's on CPU
device = torch.device("cpu")
model.to(device)

def convert_python_to_r(python_code):
    # Prompt to guide the model
    prompt = f"""### Task:
Convert the following Python code to equivalent R code.

### Python code:
{python_code}

### R code:
"""

    # Tokenize input
    input_ids = tokenizer(prompt, return_tensors="pt", truncation=True).input_ids.to(device)

    # Generate
    outputs = model.generate(
        input_ids,
        max_length=512,
        temperature=0.2,
        do_sample=True,
        pad_token_id=tokenizer.eos_token_id
    )

    # Decode result
    result = tokenizer.decode(outputs[0], skip_special_tokens=True)

    # Extract R code from the result (after prompt)
    if "### R code:" in result:
        result = result.split("### R code:")[-1].strip()

    return result

# Gradio interface
gr.Interface(
    fn=convert_python_to_r,
    inputs=gr.Textbox(lines=10, placeholder="Paste your Python code here..."),
    outputs="text",
    title="Python to R Code Converter",
    description="Converts Python code to R using Replit Code Model (3B). Optimized for Hugging Face CPU Basic tier."
).launch()