Most Popular Posts Of 2016
JeanFrancoisPuget 2700028FGP Visits (13436)
I wish you, all my readers, your families, and your friends, all the best for 2017. Your renewed interest made me write 34 entries in 2016.
This year I focused mostly on Python and Machine Learning, reflecting my role at IBM as the technical lead for machine learning offerings. This shows in the topics of my most popular posts for 2016:
What surprises me most is the popularity of posts about installing a specific machine learning package called xgboost. This package, and other implementation of gradient boosted machines (GBM) are becoming the tool of choice to learn from structured and semi structured data, while deep learning packages shine on unstructured data.
Readership has significantly increased over the year, as shown by this curve.
The vertical bumps are when some of my posts made the top of HackerNews list.
I am now back at optimization at IBM, in addition to machine learning. Therefore one can expect more optimization content next year. I will still cover Python and machine learning. Stay tuned.