This is an open source implementation of the baselines presented in the Amazon Product Search KDD CUP 2022.
We launched the baselines experiments creating an environment with Python 3.6 and installing the packages shown below:
pip3 install requirement.txt
Before to launch the script below, it would be necessary to login in aicrowd using the Python client aicrowd login.
The script below downloads all the files for the three tasks using the aicrowd client.
cd data/
bash download-data.shNote : After download data, please move these data to task1, task2.
For task 1, we fine-tuned roberta-xlm models for each query_locale.
We used the query and title of the product as input for these models.
python3 main.py
For task 1, we pre-train roberta-xlm models for each product_title + product_bullet_point.
We used the query and title of the product as input for these models.
python3 maskLMing.py
BTW, we not yet submit pretrain model to leaderboard.