Sitemap & RSS Feed Tags

Learn about data science in real life and machine learning in production

Issue 13: Kubernetes and DevOps concepts

May 08, 2021 Kubernetes is a tool. The concepts that are around this one are interesting. Zero downtime deployment, containers orchestration, scaling, etc. They help to deploy safely and then to get solid ML systems. It's relevant to understand these concepts. This week, I would like to highlight some of them through different resources. ...

Kubernetes - Be sure of being ready for real zero downtime deployment

May 05, 2021 Readiness is an important concept in Kubernetes that avoids you getting a temporary error when deploying. It's a key to get real zero downtime deployment. ...

Issue 12: Bayesian AB Tests

Apr 20, 2021 Recently, for an R&D project, I had to implement Bayesian AB tests. As AB tests are an important key to develop safely and surely, I decided to present to you what I've learnt so far. I focused my reasearch on the Pymc3 library. ...

Issue 11: Machine Learning Design Patterns

Mar 25, 2021 ML design patterns for your ML journey. ...

Issue 10: Challenges in Deploying Machine Learning

Mar 15, 2021 I'm glad to see that we can find more and more papers about deploying Machine Learning. Challenges are multiple and everywhere. Let's dive a bit into these challenges and resources that discuss these. ...

Different ways to tackle the data labelling bottleneck in machine learning

Mar 09, 2021 Data are the food of machine learning training. There are more and more data everyday. But most of the time, these data are unlabelled. Labelling them manually is expensive and boring. There are different ways to tackle this problem. Active learning Active learning optimises labelling. It extracts the data that must be labelled. The system requests a manual labelling for identified cases. Those depend on the strategy you choose. I will cover only two of these strategies. ...

Issue 9: Code standardization, container orchestration, lakehouse, cats: concepts needed to productionalize machine learning models

Mar 01, 2021 Code standardisation with Pylint, container orchestration with Kubernetes, lakehouse with DeltaLake, working with cats <3, these are many concepts that can be useful to productionalize machine learning. This is what I saw recently and thought interesting. ...

Machine Learning Design Patterns

Feb 23, 2021 I would like to tell you a story. I was trying to understand a bug. I spent maybe 5 pomodoros on it. Pomodoro is a technique that helps you to stay focus. I couldn't find the root cause of my bug. I was quite frustrated but still motivated. I decided to stop my session of pomodoros and left my work for the day. I went to my beautiful cat and started petting her. I was starting relaxing when at a sudden I got the solution to my bug. ...

Your cat might be better than focus

Feb 23, 2021 I would like to tell you a story. I was trying to understand a bug. I spent maybe 5 pomodoros on it. Pomodoro is a technique that helps you to stay focus. I couldn't find the root cause of my bug. I was quite frustrated but still motivated. I decided to stop my session of pomodoros and left my work for the day. I went to my beautiful cat and started petting her. I was starting relaxing when at a sudden I got the solution to my bug. ...

Issue 8: Code standardisation and MLOps seen by Databricks

Feb 16, 2021 Code standardisation inside the same team is important. There are different ways to achieve this goal. In this newsletter, I also would like to talk about MLOps and the way Databricks presents that. ...

Issue 7: Recommendation systems and data engineering jobs

Feb 03, 2021 Once, I've assessed different recommendation systems through AB tests. They were black boxes for me. I've decided to focus my recent techno watch on recommendation systems. I wanted to understand better these black boxes. There are many ways to do recommendation systems. The solutions depend on your context and needs. In this newsletter, I also would like to talk a bit about the roles of data engineers or machine learning engineers. ...

How can you connect to MLFlow registry remotely?

Jan 25, 2021 MLFlow is helpful when you are looking for reproducibility and MLOps. This tutorial will lead you to connect to MLFlow model registry from outside. ...