---
title: Playing Card Predictor
emoji: 🃏
colorFrom: blue
colorTo: purple
sdk: gradio
app_file: app.py
pinned: false
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

# Playing Card Face/Value Prediction App

This Gradio application uses an AutoGluon MultiModalPredictor to classify images of playing cards, identifying whether the image shows the "Face" or the "Value" of the card.

## How to Use

1.  **Upload an Image:** Use the "Input image" section to upload a photo of a playing card. You can upload a file from your computer, use your webcam, or paste an image from your clipboard.
2.  **View Processed Image:** The "Preprocessed image" section will display a resized version of your input image, showing how the model receives the image for processing.
3.  **See Prediction Results:** The "Prediction Status" will indicate if the prediction was successful. The "Class probabilities" section will show the model's confidence scores for the "👑 Face" and "🔢 Value" classes. The class with the highest probability is the model's prediction.

## Model Information

This app utilizes an AutoGluon MultiModalPredictor trained on a dataset of playing card images. The model is designed to distinguish between images focusing on the face cards (King, Queen, Jack, Ace) and those showing the numerical value cards (2 through 10).

The trained model is sourced from the Hugging Face Hub repository: `jennifee/nnl_automm_model`.

## Dataset

The model was trained on a custom dataset of playing card images, specifically curated for the task of classifying images based on whether they depict a face card or a value card.

## About the Creator

This app was created as a demonstration of deploying AutoGluon models with Gradio on Hugging Face Spaces.