app.py

from huggingface_hub import InferenceClient
import pandas as pd
import re
import random
import csv
import tempfile
import gradio as gr
client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")

def extract_sentences_from_excel(file):
    df = pd.read_excel(file)
    sentences = []
    for row in df.values.tolist():
        text = ' '.join(str(x) for x in row)
        new_sentences = re.split(r'(?<=[^.!?])(?=[.!?])', text)
        sentences.extend([s.strip() for s in new_sentences if s.strip()])
    return sentences

def generate_synthetic_data(file, temperature, max_new_tokens, top_p, repetition_penalty):
    sentences = extract_sentences_from_excel(file)
    random.shuffle(sentences)

    generated_data = []

    for sentence in sentences:
        sentence = sentence.strip()
        if not sentence:
            continue

        generate_kwargs = {
            "temperature": temperature,
            "max_new_tokens": max_new_tokens,
            "top_p": top_p,
            "repetition_penalty": repetition_penalty,
            "do_sample": True,
            "seed": 42,
        }

        try:
            output = client.generate(sentence, **generate_kwargs, return_full_text=True)
            synthetic_data = output.text.strip()

            generated_sentences = re.split(r'(?<=[\.\!\?:])[\s\n]+', synthetic_data)
            generated_sentences = [s.strip() for s in generated_sentences if s.strip() and s != '.']

            for generated_sentence in generated_sentences:
                generated_data.append({'Original Sentence': sentence, 'Synthetic Data': generated_sentence})

        except Exception as e:
            print(f"Error generating data for sentence '{sentence}': {e}")

    with tempfile.NamedTemporaryFile(mode='w', newline='', delete=False, suffix='.csv') as tmp:
        fieldnames = ['Original Sentence', 'Synthetic Data']
        writer = csv.DictWriter(tmp, fieldnames=fieldnames)
        writer.writeheader()

        for data in generated_data:
            writer.writerow(data)

        tmp_path = tmp.name

    return tmp_path

gr.Interface(
    fn=generate_synthetic_data,
    inputs=[
        gr.File(label="Upload Excel File", file_count="single", file_types=[".xlsx", ".xls"]),
        gr.Slider(label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs"),
        gr.Slider(label="Max new tokens", value=256, minimum=0, maximum=5120, step=64, interactive=True, info="The maximum numbers of new tokens"),
        gr.Slider(label="Top-p (nucleus sampling)", value=0.95, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens"),
        gr.Slider(label="Repetition penalty", value=1.0, minimum=1.0, maximum=2.0, step=0.1, interactive=True, info="Penalize repeated tokens"),
    ],
    outputs=gr.File(label="Synthetic Data CSV"),
    title="Synthetic Data Generation",
    description="Generate synthetic data from sentences in an Excel file and save it to a CSV file.",
    allow_flagging="never",
).launch()