dataset preparation

[joaquinsalas]

Hello. Thanks for making your code available.
Would you please tell me how should I prepare my dataset (currently a CSV file) to use your code?

Thanks, Joaquin

[joshua-shuhan]

Hi! Thanks for your interest in our work. The basic guideline is as follows.

1. First, it is better to create a folder to store the data (e.g., folder data_{dataset_name} in the repo).
2. You can then create a py file to process the dataset (e.g., dataset_breast.py, dataset_census_onehot.py in the repo). If the dataset only contains numerical variables, I believe the dataset_breast.py file is a good example to follow.
3. Finally, create a file to run the experiment (e.g., exe_breast.py in the repo).

Most preparation works can be done by copy and paste. But be careful about missing values handling (e.g., Line 12 in dataset_breast.py) and whether your dataset contains some columns that need to be dropped beforehand.

Don't hesitate to ask if you have any further questions : )