I am a new one in Data Science and Machine Learning. I have some experience in Java, Python, SQL, Jupyter and most polular libraries like scikit-learn, numpy, pandas, tensorflow, keras. What tools would you recommend else? And what OS is more suitable for DS (I've heard that Debian)?
A few essential packages come to mind that every Python statisticians must know:
SciPy, which is based on NumPy, extends NumPy capabilities. You can do linear algebra, probability theory, integral calculus and many more tasks with SciPy.
StatsModels - estimating of statistical models, conducting statistical tests, and statistical data exploration.
Matplotlib - 2-dimensional static plotting, from histograms and scatterplots to non-Cartesian coordinates graphs
Seaborn - high-level API based on matplotlib. It has better default settings for statistical charts, along with a rich collection of visualizations including some complex types like time series, jointplots, and violin diagrams.