Could you be a bit more specific with what you mean by visualisation for machine learning?
Yellowbrick is a (brilliant) wrapper on top of Matplotlib that provides useful predefined plots that are useful for hyperparameter tuning, algorithm choice and feature engineering.
If you want something that is more related to exploring your data before you get into the actual modelling part, then there's plenty out there. You might want to look at:
Matplotlib - the de facto plotting library for Python
Seaborn - also built on top of Matplotlib providing high level statistical plots
Bokeh - nice for making interactive charts
Altair - uses declarative grammar to build graphs which can be more intuitive and require less code