Derek Jedamski (@DSJ_stats) header
Derek Jedamski (@DSJ_stats) profile picture

Statistics. Programming. Machine Learning. Data Viz. Reformed blogger. Always learning.

Rochester, NY
Joined October 2015
Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)Photo by Derek Jedamski (@DSJ_stats)
Wow - this is super cool. Stack Overflow search built in, commenting ability, ability to easily share interactive notebooks, etc. colab.research.google.com/no…
3
Derek Jedamski retweeted
With 2018 underway, GitHub’s Data Science Team looked at the growth of contributors, visits, and stars across some of the most popular open source projects to identify trends for the year ahead. See what they found: github.com/blog/2499-open-so…
3
90
7
185
Derek Jedamski retweeted
GitHub's security alerts were named @infoworld's #4 technology of the year! infoworld.com/article/325182… So proud of @mattyoho @pmarsceill @NaomiPlasterer @robospork @shegoesbysheri @ansate @mcolyer @beneddy @bathompso @kav_gan @D_Jedamski & everyone else who contributed 😍
1
9
2
32
I got way too much enjoyment out of this.
1
2
Most of the time, AI is just a machine learning problem that hasn't been solved yet (I despise the term AI). snip.ly/bukyq#http://www.pop…
1
1
Just launched today, still a WIP but check it out and let me know what you think! github.com/dashboard/discove…
1
Derek Jedamski retweeted
Does the NFL have parity? Mostly not. Slides from my talk at #JSM2017 statsbylopez.github.io/JSM20…
1
25
5
48
This is probably pretty close to how I would've labeled it. mango-solutions.com/radar/
“Programming is hard. That’s precisely why you should learn it.” by @RogerAFCollier medium.freecodecamp.com/make…
Derek Jedamski retweeted
Distill is a interactive, visual journal for machine learning research. distill.pub/about/
1
176
23
254
you might experience moving from something like adaptive boosting to bagging?
time might be constraint on using adaBoost. Testing out BaggingClassifier wrapper for SVC now, any experience on performance degredation
1
Looking to use ensemble of SVMs but fitting on full dataset (~180k examples) is very time consuming. Originally wanted to use adaBoost but