augmentation

Inspired by imgaug:

https://github.com/aleju/imgaug

https://github.com/aleju/imgaug/blob/0101108d4fed06bc5056c4a03e2bcb0216dac326/imgaug/augmenters/meta.py#L3188

There are quite a few functions in the repo that need to be annotated.
This issue will be left open for external contributions. If you wish to contribute, please submit a PR for the same!

Thanks!
-J

Can you please make it possible to choose top_n most similar words instead of having to replace with random synonym from the default 10 synonyms.

Using the same concept used in keras segmentation example add pytorch segmentation example