Patsy (PyData Berlin)

PyData Berlin / Canada Day, 2017 at 11:45am - 12:30pm

Transcript

1. @maxhumber Canada Day, 2017
2. patsy
4. r python patsy intro data motivation ?? summary
15. type A type b
19. patsy
22. 1 0
24. create data
30. r python patsy
35. wait for it
47. let’s try again
59. not the only one …
65. Patsy is a Python library for describing statistical models (especially linear models, or models that have a linear component) and building design matrices. Patsy brings the convenience of R "formulas" to Python. https://github.com/pydata/patsy
66. pip install patsy
75. <bonus 1>
81. </bonus 1>
84. Beer: 1.5, Warm: -0.5, Family: Yes
85. Beer: -0.9, Warm: 0.3, Family: No
87. https://patsy.readthedocs.io/en/latest/quickstart.html
88. https://patsy.readthedocs.io/en/latest/quickstart.html
90. Our goal is to make Patsy the de facto standard for describing models in Python, regardless of the underlying package in use – just as formulas are the standard interface to all R packages. Therefore we’ve tried to make it as easy as possible for you to build Patsy support into your libraries. https://patsy.readthedocs.io/en/latest/library-developers.html
93. PyMC3 is a Python package for Bayesian statistical modeling and Probabilistic Machine Learning which focuses on advanced Markov chain Monte Carlo and variational fitting algorithms. Its flexibility and extensibility make it applicable to a large suite of problems. https://pymc-devs.github.io/pymc3/
97. summary
99. patsy
100. thanks! NJS https://vorpus.org/
101. maxhumber
102. maxhumber let’s talk let’s celebrate