Chevron Left
Back to Bayesian Inference with MCMC

Learner Reviews & Feedback for Bayesian Inference with MCMC by Databricks

3.1
stars
17 ratings

About the Course

The objective of this course is to introduce Markov Chain Monte Carlo Methods for Bayesian modeling and inference, The attendees will start off by learning the the basics of Monte Carlo methods. This will be augmented by hands-on examples in Python that will be used to illustrate how these algorithms work. This will be the second course in a specialization of three courses .Python and Jupyter notebooks will be used throughout this course to illustrate and perform Bayesian modeling with PyMC3. The course website is located at https://sjster.github.io/introduction_to_computational_statistics/docs/index.html. The course notebooks can be downloaded from this website by following the instructions on page https://sjster.github.io/introduction_to_computational_statistics/docs/getting_started.html. The instructor for this course will be Dr. Srijith Rajamohan....
Filter by:

1 - 3 of 3 Reviews for Bayesian Inference with MCMC

By GBM

Dec 29, 2021

I had to look in other materials to understand most of the algorithms, the way it was presented just reading the material doesn't help much. As a result, I ended up preferring to just read the material rather than watch it. So, I think it could be a more dynamic class to improve didactics.

By Ross

Mar 7, 2022

This course is inappropriately calibrated. At best, it is a fair refresher for someone who has already taken a graduate level course in Bayesian statistics. The coding assignments are do-able, but the course content does not include an example solution to the ungraded coding homework assigments. The lectures are fair to poor, with notes that neither visually convey the intuition of each method nor the mathematical details. In summary, the class is an ok 8 hour refresher for someone who already knows Bayesian statistics and MCMC.

By Randall B

Dec 13, 2022

The presenter simply read the notebooks, didn't add anything. He seemed bored most of the time.

Showed an example where the predicted distributions were consistent but didn't show one where they were not consistent. There were many concepts in the course that would benefit from more demonstrations of how the algorithms work, rather than describing them.

I was disappointed with this course and will avoid courses from Databricks in the future.