<!-- TITLE: Probabilistic programming and Bayesian statistics --> <!-- SUBTITLE: --> # The easy way: PyMC3 + [PyMC3](https://docs.pymc.io/) is so freaking easy to use. + Don't use your laptop processor. + Spin up an EC2 instance in AWS + Search for a "data science" AMI when it asks + Once it's running and you're connected, run a Jupyter notebook to run and modify scripts + This way you don't have to worry about compilation issues on your specific processor, setting up GPU-access, etc. + [here's a stellar tutorial](https://docs.pymc.io/notebooks/Diagnosing_biased_Inference_with_Divergences.html), and if that's too high level for you, work through the details of the [deeper (longer) explanation](https://towardsdatascience.com/estimating-probabilities-with-bayesian-modeling-in-python-7144be007815), or dive into [Expectation Maximization (EM) algorithm](https://people.duke.edu/~ccc14/sta-663/EMAlgorithm.html), or [this one](https://towardsdatascience.com/maximum-likelihood-estimation-how-it-works-and-implementing-in-python-b0eb2efb360f) + [e.g. Dirchelet!!](https://docs.pymc.io/api/distributions/multivariate.html#pymc3.distributions.multivariate.Dirichlet)