written by Eric J. Ma on 2019-04-29 | tags: pycon python data science conferences
I'm headed out to PyCon 2019! This year, I will be co-instructing two tutorials, one on network analysis and one on Bayesian statistics, and delivering one talk on Bayesian statistics.
The first... (read more)
(395 words, approximately 2 minutes reading time)written by Eric J. Ma on 2019-03-24 | tags: data science machine learning
Have you heard of variance explained as a loss function and machine learning metric? Turns out it’s quite useful and interpretable. I’d like to share this new learning with you.
Read on... (499 words, approximately 3 minutes reading time)written by Eric J. Ma on 2019-03-22 | tags: python hacks tips and tricks data science productivity coding
In praise of `functools.partial`, and how I used it in a Flask/Bokeh app!
Read on... (420 words, approximately 3 minutes reading time)written by Eric J. Ma on 2019-03-20 | tags: data science productivity
My tooling, routines, and techniques for getting things done and learning new things!
Read on... (846 words, approximately 5 minutes reading time)written by Eric J. Ma on 2019-03-01 | tags: data science programming best practices
In this Q&A-style blog post, I detail how data scientists can begin to engage in pair coding as a more common practice in our day-to-day work, and why we should spend the time to do it as much as we can afford.
Read on... (865 words, approximately 5 minutes reading time)written by Eric J. Ma on 2019-01-28 | tags: data science data products minimum viable products
I would like to encourage you to build more "minimum viable products" of your projects. Come learn why they’re so valuable!
Read on... (102 words, approximately 1 minute reading time)written by Eric J. Ma on 2019-01-21 | tags: scalability bayesian model dose response parameter learning model specification convergence shrinkage large dataset nuts mcmc advi variational inference neural networks random sampling biochemistry data modeling
I've been exploring a Bayesian hierarchical 4-parameter dose response model at work. Initially, I used a few thousand samples for prototyping, but I've now scaled up to 400K+ samples. Fitting the model with NUTS would've taken a week, but ADVI did the job in just 2.5 hours. 🚀 This experience has given me a new appreciation for ADVI, even in simpler models with large datasets. 🧠
Read on... (365 words, approximately 2 minutes reading time)written by Eric J. Ma on 2018-12-25 | tags: data science conda hacks
The conda package manager has, over the years, become an integral part of my workflow. I use it to manage project environments, and have built a bunch of very simple hacks around it that you can adopt too. I'd like to share them with...
(read more)
written by Eric J. Ma on 2018-12-16 | tags: data science bayesian
Here are my notes from learning about Gaussian Processes. It's been a long intellectual journey; hope you find my notes useful.
Read on... (315 words, approximately 2 minutes reading time)written by Eric J. Ma on 2018-12-09 | tags: deep learning bayesian math data science
Last week, I picked up Jeremy Kun's book, "A Programmer's Introduction to Mathematics". In it, I finally found an explanation for my frustrations when reading math papers:
What programmers would consider "sloppy" notation is one... (read more)
(777 words, approximately 4 minutes reading time)