So tonight is going to be the 13th Hors Série of Season 4 of the Paris Machine Learning meetup. This will bring about to a record 23 the number of meetups Franck and I have made happen this season!
Tonight will be a presentation of two toolboxes around the issue of Automated Machine Learning. One toolbox is by Axel de Romblay and the other is by Alexis Bondu.
There is limited seating however we will have a streaming (see below).
Thank to Sewan for hosting us. Here is the video for the streaming:
- 0 - Preview of the slides
- 1 - Introduction
- How to ?
- What's important ?
- 2 - Theories and Baseline to automate Machine Learning
- Overview of the different approaches to automate Machine Learning (bayesian, ...)
- 3 - Demo & Coding
- Edge ML & MLBox
- Comparison of these tools with the same dataset
- 3-1 - MLBox (Machine Learning Box)
- MLBox is a powerful Automated Machine Learning python library. It provides the following functionalities:
- - Fast reading and distributed data preprocessing / cleaning / formatting
- - Highly robust feature selection and leak detection
- - Accurate hyper-parameter optimization in high-dimensional space
- - State-of-the art predictive models for classification and regression (Deep Learning, Stacking, LightGBM,...)
- - Prediction with models interpretation
- To learn more about this tool and how to get it installed, please refer to:
- 3.2 - Edge ML
- Edge ML is an Automated Machine Learning software which implements the MODL approach, which is unique in two respects :
- i) it is highly scalable and avoids empirical optimization of the models by grid-search;
- ii) it provides accurate and very robust models.
- To participate in the interactive demonstration, you can install Edge ML :
- 1 - Install the shareware version: link
- 2 - Create and download your licence file: link
- 3 - Copy / Paste your licence file to your home directory
- 4 - Download the Python wrapper (with demo example): link
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.