app.py

import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, set_seed, pipeline


title = "Python Code Generator"
description = "This is a space to convert English text to Python code using the [codeparrot-small-text-to-code](https://huggingface.co./codeparrot/codeparrot-small-text-to-code) model, a pre-trained Python code generation model trained on a dataset of docstrings and Python code extracted from Jupyter notebooks available at [github-jupyter-text](https://huggingface.co./datasets/codeparrot/github-jupyter-text)."
example = [
    ["Utility function to calculate the precision of predictions using sklearn metrics", 65, 0.6, 42],
    ["Let's implement a function that calculates the size of a file called filepath", 60, 0.6, 42],
    ["Let's implement the Bubble Sort sorting algorithm in an auxiliary function:", 87, 0.6, 42],
    ["Function to calculate the nth Fibonacci number.", 65, 0.6, 42],
    ["Function to calculate the factorial of a number.", 65, 0.6, 42],
    ["Function to reverse a string.", 65, 0.6, 42],
    ["Function to check if a number is prime.", 65, 0.6, 42],
    ["Function to generate the Fibonacci sequence up to the nth term.", 65, 0.6, 42],
    ["Function to generate the factorial sequence up to the nth term.", 65, 0.6, 42],
]


# Change the model to the pre-trained model
tokenizer = AutoTokenizer.from_pretrained("codeparrot/codeparrot-small-text-to-code")
model = AutoModelForCausalLM.from_pretrained("codeparrot/codeparrot-small-text-to-code")

def create_docstring(gen_prompt):
    return "\"\"\"\n" + gen_prompt + "\n\"\"\"\n\n"

def validate_inputs(gen_prompt, max_tokens, temperature, seed):
    # Add validation logic here
    if not gen_prompt:
        raise ValueError("English instructions cannot be empty.")
    if max_tokens <= 0 or max_tokens > 256:
        raise ValueError("Number of tokens to generate must be between 1 and 256.")
    if temperature < 0 or temperature > 2.5:
        raise ValueError("Temperature must be between 0 and 2.5.")
    if seed < 0 or seed > 1000:
        raise ValueError("Random seed must be between 0 and 1000.")

def generate_code(gen_prompt, max_tokens, temperature=0.6, seed=42):
    validate_inputs(gen_prompt, max_tokens, temperature, seed)

    # Encode the input prompt
    input_ids = tokenizer.encode(gen_prompt, return_tensors="pt")

    # Set seed for reproducibility
    set_seed(seed)

    # Generate code tokens
    output = model.generate(
        input_ids,
        max_length=max_tokens + input_ids.shape[-1],
        temperature=temperature,
        pad_token_id=tokenizer.eos_token_id,
        num_return_sequences=1
    )

    # Decode the generated tokens into Python code
    generated_code = tokenizer.decode(output[:, input_ids.shape[-1]:][0], skip_special_tokens=True)

    return generated_code



def save_to_text_file(output_text):
    with open("generated_code.txt", "w") as file:
        file.write(output_text)

iface = gr.Interface(
    fn=generate_code, 
    inputs=[
        gr.Textbox(label="English instructions", placeholder="Enter English instructions..."),
        gr.inputs.Slider(
            minimum=8,
            maximum=256,
            step=1,
            default=8,
            label="Number of tokens to generate",
        ),
        gr.inputs.Slider(
            minimum=0,
            maximum=2.5,
            step=0.1,
            default=0.6,
            label="Temperature",
        ),
        gr.inputs.Slider(
            minimum=0,
            maximum=1000,
            step=1,
            default=42,
            label="Random seed for generation"
        )
    ],
    outputs=gr.Code(label="Generated Python code", language="python", lines=10),
    examples=example,
    layout="horizontal",
    theme="peach",
    description=description,
    title=title
)
iface.launch()