https://naoko.github.io/tags/learning/Recent content in Learning on the art of simplicityHugoen-usSun, 08 Jul 2018 00:00:00 +0000https://naoko.github.io/posts/2018-07-08-ml-platform/Sun, 08 Jul 2018 00:00:00 +0000https://naoko.github.io/posts/2018-07-08-ml-platform/<h3 id="the-problem">The problem:</h3> <p>When you start small Machine Learning team with a few projects, your experiment is done via Jupyter Notebook and maybe the notebook is in github. The notebook might contain a method to download data so it can be reproducible but it is getting harder and harder to track various experiments.</p> <p>We also need to make sure models does not train on corrupted / skewed data and only high quality model are pushed to production. Currently these processes are manual and not centralized nor has unified common tool.</p>