D2DraftNet-App

D2DraftNet is a deep-learning-based Dota 2 draft prediction tool. It analyzes hero selections and predicts the win probability for each team based on historical match data.

🚀 Features

• Hero Draft Selection: Interactive UI for selecting Radiant and Dire drafts.
• Win Probability Prediction: Uses a trained deep learning model to predict the outcome.
• User-Friendly Web Interface: Built with Flask for easy accessibility.

---

💻 How to Update the Model

D2DraftNet is a deep learning model for predicting Dota 2 draft outcomes which is used in the web app. The package can be found at PyPI. If there is a newer version available on PyPI, add it to the D2DraftNet App by running the following commands:

Remove the old version

poetry remove d2draftnet

Clear the cache.

poetry cache clear

Install the new version.

poetry add d2draftnet@latest

Verify the version matches the latest release on PyPI.

poetry show d2draftnet

---

📦 Dependencies

D2DraftNet requires the following dependencies, which are managed using Poetry:

• Python >=3.10
• Flask >=3.1.0
• Torch >=2.6.0
• Pandas >=2.2.3
• Requests >=2.32.3
• BeautifulSoup4 >=4.13.3

Install all dependencies using:

poetry install

---

💻 How to Use

1️⃣ Clone the Repository

git clone https://github.com/gkerr708/D2DraftNet-App.git
cd D2DraftNet-App

2️⃣ Set Up Poetry Environment

Make sure you have Poetry installed:

curl -sSL https://install.python-poetry.org | python3 -

Then install dependencies:

poetry install

3️⃣ Run the Web App

poetry run python app.py

Then open http://127.0.0.1:5000 in your browser.

---

🤝 Contributing

Pull requests are welcome! Please make sure your contributions follow best practices and are properly tested.