TL;DR Summary: News: my talk at PyCon Canada 2018 summarizing my thoughts for the broader context of research software development is available here. Python and its ecosystem are ideal for neuroinformatics, and we must try to get all the essential and established techniques implemented in pure python Community efforts (for testing & validation) must be …
Quantitative comparison of scikit-learn’s predictive models on a large number of machine learning datasets: A good start
This is an enormous effort with great potential for helping bioinformaticians in their research.
Are physicians worried about computers machine learning their jobs?
Reports of unintended consequences due to machine learning have been greatly exaggerated!
Open science is better science!
Repeat after me, open science is better science! I believe that, and in that spirit, I am starting to do my best to move the peer review process to open forums when possible (authors permitting of course). Here is the first one on a preprint titled: "The ANTs Longitudinal Cortical Thickness Pipeline" posted on biorxiv. My comments are …
Must read machine learning papers
Must read machine learning papers on performance evaluation and comparison.
On what kind of twitter persona to maintain: be yourself!
These are exciting times. With just 140 characters and one click, you could potentially reach millions of people. You don't need to be famous, you just need great content. May be a hashtag. A message of love. An ordinary moment of joy. Enthusiastic cheering for your team. Spontaneous expression of disappointment. You just need to join …
Continue reading On what kind of twitter persona to maintain: be yourself!
A must read perspective on “Machine Learning and Prediction in Medicine — Beyond the Peak of Inflated Expectations”
Every once in a while, there appears a paper that is so good, so well-written and so packed with insights, that I would read it a few times... that I wished I had written it myself. That happened when I read this perspective paper by Jonathan Chen and Steven Asch. Normally I'd tweet a link …
Leave one scheme out of your cross validation studies
Placeholder for an upcoming blog post to advocate against the use of leave one out cross-validation. Do not read too much into the following plots until I provide you the full context and details.
cross invalidation 