Machine Learning in R with H2O and LIME: slides, photos and video
This article is originally published at http://www.milanor.net/blog
Hello, everyone! Did you have fun at June 25th’s workshop with H2O and Lime? If you were there, we’re sure you did. If you weren’t… here’s a recap, with the explanation, the link to the live event and all the materials too!
The workshop started at 19 in Mikamai, and it began with a bang! Joe Chow, evangelist of H2O, launched us right away into the world of Machine Learning. He presented to us AutoML, a framework for model’s training and comparison, which is able to fit and compare different model families with the use of few parameters. AutoML can be used for automating the machine learning workflow, which includes automatic training and tuning of many models within a user-specified time-limit. After introducing AutoML, Joe showed us LIME, a package for model interpretation.
If you want to know more, click on the link below to look at the slides or check the Github repo.
by Jo-fai (Joe) Chow (Github repo)
Then, after the theory, we got practical. We met Moneyball, an app to detect the most promising baseball players that got payed less - the ones that’s worth investing on, basically. You can find both the slides and the demo below. The developers used different models as examples, both dataset free and proprietary, and then they built a shiny app to make it more easy for business experts - the ones who were going to buy the players.
If you're more a listening type than a reading type of person, the full recording is available on our Facebook channel:
— Jo-fai (Joe) Chow (@matlabulous) 11 luglio 2018
With full stomaches and full minds, the only thing we could do was have a chat together. We can’t thank Joe, H2O and Quantide enough for offering us this huge opportunity.
The post Machine Learning in R with H2O and LIME: slides, photos and video appeared first on MilanoR.
Please visit source website for post related comments.