Will Charles
Will Charles
Commenting out / removing these two lines seems to work and ensures the split ratio is maintained.  It ensures that the data is not duplicated in test and train....
> Thank you [@dannycg1996](https://github.com/dannycg1996) , [@drwillcharles](https://github.com/drwillcharles) . For classification, we want to make sure the labels are complete in both training and validation data, thus we'll concat the first instance...