README.md

metadata

title: Pokémon Price Predictor
emoji: 🃏
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 4.38.1
app_file: app.py
pinned: false
license: mit
tags:
  - pytorch
  - scikit-learn
  - gradio
  - machine-learning
  - tabular-classification
  - price-prediction
  - finance
  - pokemon
  - pokemon-cards
  - tcg
  - collectibles

PokePrice: Pokémon Card Price Trend Predictor

This application uses a PyTorch-based neural network to predict whether the market price of a specific Pokémon card will rise by 30% or more over the next six months.

How It Works

1. Enter a Card ID: Input the numeric TCGPlayer ID for a specific Pokémon card. You can find this ID in the URL of the card's page on the TCGPlayer website (e.g., tcgplayer.com/product/84198/...).
2. Get Prediction: The model analyzes various features of the selected card, such as its rarity, type, and historical price data, to make a prediction.
3. View Results: The application displays:
   • The card's name and the prediction (whether the price is expected to RISE or NOT RISE).
   • The model's confidence level in the prediction.
   • A direct link to view the card on TCGPlayer.com.
   • The actual historical outcome if it exists in the dataset, for comparison.

The Technology

• Model: A simple feed-forward neural network built with PyTorch.
• Data: The model was trained on a custom dataset derived from the Pokémon TCG API and historical market data from TCGPlayer.
• Frontend: The user interface is created with Gradio.
• Deployment: Hosted on Hugging Face Spaces.