Split data into Train, Test and Validation Data

[Pomo (Malcolm McLean)]

Hi Jay. Typically an unlabelled Test set is provided by the machine leaning practitioner, or by the problem statement, or by the competition. Please see here for options for adding a Test set to the DataBunch:

Using a labeled test set with fastai?

This question has been asked and answered already… See here how to create a second DataBunch when you want to validate you model on another set. As for the F1 score, it is available on non-multi-class data if you use FBeta(beta=1) (another question that has been asked a lot on the forum). Why does it make sense to have no labels for your test set? Well, when you are in a Kaggle competition, your test set doesn’t have labels and you need to provide predictions. The fastai test set is there for …

If you want to split out your own Test set, IMHO the easiest way is to learn how to use DataFrames (Pandas). With DataFrames you can load in the csv, take a sample, write a new CSV, move files, etc. Then fastai will split the remaining set into Training and Validation.