Generating Personalized Recipes from Historical User Preferences

This is our PyTorch implementation for the paper:

Generating Personalized Recipes from Historical User Preferences, EMNLP 2019

The code is tested on a Linux server (with NVIDIA GeForce Titan X Pascal / NVIDIA GeForce GTX 1080 Ti) with PyTorch 1.1.0 and Python 3.6.

Requirements

• Python 3
• Pytorch v1.0+

Data

Backing data can be found on Kaggle.

Running Models

To train a model, see the recipe_gen/models/<model>/train.py file for that particular model (Baseline train.py linked). Likewise, run the test.py in the folder with arguments as listed to evaluate.

Citation

If you find this repository useful for your research, please cite our paper:

@inproceedings{majumder-etal-2019-generating,
    title = "Generating Personalized Recipes from Historical User Preferences",
    author = "Majumder, Bodhisattwa Prasad  and
      Li, Shuyang  and
      Ni, Jianmo  and
      McAuley, Julian",
    booktitle = "EMNLP",
    year = "2019",
    url = "https://www.aclweb.org/anthology/D19-1613",
    doi = "10.18653/v1/D19-1613",
    pages = "5975--5981",
}