General:
    Dive into Python 3: http://www.diveintopython3.net/index.html. Very useful to me when I started learning python.
    Software carpentry: https://software-carpentry.org/. They have lessons in many other topics beside python.
    Code academy: https://www.codecademy.com/learn/python
    Many EdX and Coursera courses, often introductory CS courses which can teach other useful skills (algorithms and datastructures)

Reference and help:
    Just google python + description of problem
    Usually answer is on stackoverflow
    The python docs: https://docs.python.org/3/ (of course) 

Youtube channels with talks / tutorials:
    Pycon, e.g.:  https://www.youtube.com/channel/UCwTD5zJbsQGJN75MwbykYNw (2016), https://www.youtube.com/channel/UCgxzjK6GuOHVKR_08TT4hJQ (2015)
    PyData (various conferences each year): https://www.youtube.com/user/PyDataTV and other per-conference channels
    Scipy conferences, e.g. https://www.youtube.com/playlist?list=PLYx7XA2nY5Gf37zYZMw6OqGFRPjB1jCy6
    Enthought: https://www.youtube.com/user/EnthoughtMedia
    Continuum Analytics: https://www.youtube.com/channel/UCND4vKhJssAtK8p1Blfj14Q
    
Some talks I liked and/or learned a lot from:
    Ned Batchelder, "Facts and Myths about Python names and values" https://www.youtube.com/watch?v=_AEJHKGk9ns
    Thomas Ballinger, "Finding closure with closures" https://www.youtube.com/watch?v=E9wS6LdXM8Y
    David Baumgold, "Advanced Git" https://www.youtube.com/watch?v=4EOZvow1mk4
    David Beazley, "Discovering python" https://www.youtube.com/watch?v=RZ4Sn-Y7AP8.

Three cool big libraries I didn't get a chance to mention elsewhere:
    Scikit-learn (machine learning): http://scikit-learn.org/
    Sympy (symbolic math): http://www.sympy.org/en/index.html
    Cython (C/C++ extensions): http://cython.org/

Three cool small libraries I didn't get a chance to mention elsewhere:
    emcee and corner: http://dan.iel.fm/emcee/current/ MCMC sampler, great for fitting if you don't mind writing your own likelihood function
    argparse (standard library): https://docs.python.org/3/library/argparse.html Add command-line interfaces to your programs
    multihist https://pypi.python.org/pypi/multihist: convenience functions for multidimensional histograms (actually just shameless advertising for my own code)
