ADRec

Quick Start

1. Install the dependencies by running the following command:

git checkout dev
uv sync

2. Run the following command to start training the model with a specified category:

CUDA_VISIBLE_DEVICES=0 uv run main.py --category=Sports_and_Outdoors

Available categories:

• Sports_and_Outdoors
• Beauty
• Toys_and_Games
• CDs_and_Vinyl

Note that:

1. The datasets will be automatically downloaded once the category argument is specified.
2. All hyperparameters can be specified via command line arguments. Please refer to:
   • genrec/default.yaml
   • genrec/datasets/AmazonReviews2014/config.yaml
   • genrec/models/TigerDiff/config.yaml

We provide the default training script train_tigerdiff_cd.sh, train_tigerdiff_sports.sh, train_tigerdiff_beauty.sh, and train_tigerdiff_toys.sh for training the model on different categories.

Important

TODO: Tune the hyperparameters.