ToonSuggest is a simple, lightweight, and interactive movie recommender system designed for animation and cartoon titles using IMDb datasets. Built using Python and Tkinter, it allows you to train a recommendation model and get suggestions for similar titles based on textual similarity and user ratings.

Actions

1. Download 🗃️ Windows EXE File and 🗃️ Model PKL File ~v1.0

2. Open GitHub Repository source for this project.

📌 Features

• GUI-based interface with file selection, training, and recommendation
• Uses TF-IDF vectorization for content-based filtering
• Compatible with IMDb .tsv.gz data files
• One-click .exe build available (see Releases)
• No external API or internet access required

📌 Getting Started

1. Clone the Repository

git clone https://github.com/s-m-quadri/toon-suggest.git
cd toon-suggest

2. Install Dependencies

Install the required Python libraries using:

pip install -r requirements.txt

3. Run the App

python main.py

Or download the pre-built .exe from the Releases  section to run without Python.

📌 How It Works

1. Train the Model

   • Launch the training window and select:

     • title.basics.tsv.gz
     • title.ratings.tsv.gz

   • Click “Start Training” to preprocess and vectorize the data.

   • A trained model.pkl file is generated and saved.

2. Get Recommendations

   • Launch the main window.
   • Load the trained model.
   • Enter the name of a movie and click “Get Recommendations”.

📌 Project Structure

ToonSuggest/
├── main.py                  # Entry point for the GUI application
├── model.pkl                # Sample trained model (for demo/testing)
├── setup.spec               # PyInstaller config for .exe build
├── requirements.txt         # Python dependencies
└── src/
    ├── gui_home.py          # Main recommendation GUI
    ├── gui_train.py         # Model training GUI
    ├── recommend.py         # TF-IDF recommendation logic
    └── train.py             # Data loading, training, and saving logic

📌 Executable Build

To create a standalone .exe:

pyinstaller --onefile --noconsole setup.spec

Or use the ready-built binary available in Releases .

📌 IMDb Dataset Download

ToonSuggest requires two official IMDb dataset files:

1. title.basics.tsv.gz – Contains title names, genres, and years
2. title.ratings.tsv.gz – Contains user ratings and number of votes

Download both from the official IMDb dataset portal:

👉 https://datasets.imdbws.com/ 

Steps:

• Visit the link above
• Download:
  • title.basics.tsv.gz
  • title.ratings.tsv.gz
• Keep both files easily accessible for selection during model training

Note: IMDb datasets are periodically updated, so you may retrain your model when new data becomes available.

📌 Walkthrough: How to Use the App

Home Window (main.py → gui_home.py)

This is the main user interface for movie recommendations.

• Title Input Field: Start typing an animated movie title to search
• Train Model: Opens the training window to build or rebuild the model
• Refresh Model: Reloads an already trained model from model.pkl
• Get Recommendations: Fetches similar movies to the selected title

The Get Recommendations button is only enabled when a valid model is loaded.

Training Window (gui_train.py)

Accessed via the “Train Model” button on the Home screen.

• Choose IMDb Files:
  • Choose title.basics.tsv.gz
  • Choose title.ratings.tsv.gz
• Reset: Clears selected files and allows re-selection
• Start Training:
  • Enabled only when both files are selected
  • Builds the model using TF-IDF vectorization and saves it as model.pkl
  • Automatically closes and updates the Home screen once training completes

During training, all inputs are disabled to avoid interruptions.

📌 License

This project is licensed under the GNU General Public License.

📌 Contact

For bug reports or feature requests, please contact: dev.smq@gmail.com