app.py

import torch
import gradio as gr
from transformers import pipeline

from typing import Dict


def food_not_food_classifier(text: str) -> Dict[str, float]:
    # Create the classifier pipeline
    food_not_food_classifier_pipeline = pipeline(
        task="text-classification",
        model="joadithya/learn_hf_food_not_food_text_classifier-distilbert-base-uncased",
        batch_size=32,
        device="cuda" if torch.cuda.is_available() else "cpu",
        top_k=None  # Returning all possible labels for a given input
    )

    # Get the outputs from the pipeline
    outputs = food_not_food_classifier_pipeline(text)[0]

    # Format output for Gradio
    output_dict = {}
    for item in outputs:
        output_dict[item["label"]] = item["score"]

    return output_dict


description = """
A text classifier model to determine whether a caption is about food or not food.

Fine-tuned from [DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased) on a [small dataset of food and not food captions](https://huggingface.co/datasets/mrdbourke/learn_hf_food_not_food_image_captions)
"""

demo = gr.Interface(
    fn=food_not_food_classifier,
    inputs="text",
    outputs=gr.Label(num_top_classes=2),
    title="Food Caption Classifier",
    description=description,
    examples=[["Nothing beats the taste of home"],
              ["Love served on a plate"],
              ["A toast with cherry on top"]] 
)

if __name__ == "__main__":
    demo.launch()