Issue
When splitting the data into the test and training data what split size should I choose?
I was told that set by default is 80 % for training data,and 20 % for test data, can somebody ensure me?
Thanks
Solution
The default split size in sklearn.model_selection.train_test_split is 75% training and 25% testing data split. Data splits should depend on the task at hand.
The rule of thumb is that you have enough representative data that represent your problem to train a model and a enough representative data to validate your trained model.
Answered By – Prayson W. Daniel
This Answer collected from stackoverflow, is licensed under cc by-sa 2.5 , cc by-sa 3.0 and cc by-sa 4.0