Due to GitHub's file size limitations, some important files are stored locally only:
dataset/train.csv(~50MB) - Main training datasetdataset/test.csv(~25MB) - Test dataset for predictions
How to get them: These are the original competition datasets. Download from the source or contact repository owner.
image_cache/*.jpg(2,183 files, ~500MB total) - Product images for computer vision- Only
image_cache/image_features.jsonis included (metadata)
How to get them: Run the download script:
python scripts/download_computer_vision_images.py- β All Python code and models
- β
Best model predictions (
results/test_out_meta_learning.csv) - β
Image metadata (
image_cache/image_features.json) - β Sample datasets for testing
- β Complete documentation
- β All requirements and setup files
git clone https://github.com/abbinavv/AmazonML.git
cd AmazonML
python -m venv .venv
source .venv/bin/activate # (.venv\Scripts\activate on Windows)
pip install -r requirements.txt
# Optional: Download images for computer vision
python scripts/download_computer_vision_images.pyThe repository includes the best model predictions (50.12% SMAPE) in:
results/test_out_meta_learning.csv- Meta-learning model (best)results/test_out_neural_enhanced.csv- Neural enhanced model