Review on FlinkML by Larry Morris

I like how easy fluml makes data preparation when using machine learning models such as Random Forests or Deep Neural Networks with TensorFlow backend! It's very nice because most other libraries require you create some python files which are not always possible if your dataset contains large amounts of categorical variables / columns etc... There isn't much more than what i mentioned above but its good enough so far :) If there were things they could improve maybe adding better documentation? Maybe also add examples where people can see exactly their input format should be formatted by this tool before putting into production - This would really helpful since sometimes formatting errors occur even after validating all inputs/columns used within DataFrameWriter class!