search

Ritam-Guha / Py_FS

A Python Package for Feature Selection
MIT License
57 stars 38 forks source link

Make test size a keyword argument #20

Closed pmcarpan closed 3 years ago

pmcarpan commented 3 years ago

In many of the algorithms, something similar to the below line has been used to split the data into two partitions.

train_test_split(train_data, train_label, stratify=train_label, test_size=0.2)

I am suggesting, instead of keeping the test_size parameter fixed to 0.2, a keyword argument can be added with a default value of 0.2. This would allow users to pass in custom values if they want while keeping a suitable default value.

Note: Please close this issue if I am mistaken.

Ritam-Guha commented 3 years ago

Thanks a lot for your recommendation. We will take care of it and let you know about the update.

Ritam-Guha commented 3 years ago

Hi @pmcarpan, I have changed the code and now a prompt will appear while running the code asking the user to enter the validation percentage. Sorry for the delay. Thanks a lot for your patience. 🙂