Feature request : Add option on the preprocessing to chose a custom text preprocessor.

[espoirMur]

What we have in the preprocessing step is already a good starting point, but we can do better by adding an option to define someone's custom preprocessing pipeline to handle what is not yet handled in the current preprocessing.

[drob-xx]

I'm wondering if you have more specifics on what you would like here and what the goals are? I took a quick look and it seems that you provide a pretty good pre-processing module and that if someone wanted to do it differently they could simply skip your pre-processor entirely and just provide a processed dataset. Am I missing something?