Here I have structured my articles according to the different categories they belong to. They do not represent an exhaustive Machine Learning curriculum but they are topics of interest I came across through my Data Science journey. Python and Machine Learning Initiation This article is a good start for anyone who would like to give…

Google Cloud Platform keeps on releasing exciting features within its Data suite. Solving one of the top challenges of Data teams, ensuring a good quality of ingested and exposed Data, is becoming simpler. — Whether in a centralized or data-mesh architecture, creating and maintaining Data pipelines over time while ensuring a good level of reliability is only the tip of the iceberg. Data applications, especially the ones embedding Machine Learning, tend to fail for numerous reasons, one of the most frequent being data drift…

Google’s Datawarehouse offers amazing ways to restrict access to the information contained in tables and views, on both vertical and horizontal dimensions, let’s discover how to set this up. — This article was originally published in Towards Data Science. Managing a data warehouse implies that multiple users, with different roles and rights, should be able to query the data they are looking for. You might decide to build one specific table for each role but this would be time and…

Google’s Datawarehouse offers amazing ways to restrict access to the information contained in tables and views, on both vertical and horizontal dimensions, let’s discover how to set this up. — Managing a data warehouse implies that multiple users, with different roles and rights, should be able to query the data they are looking for. You might decide to build one specific table for each role but this would be time and resource-consuming, not to mention the difficulties to maintain this…

A step-by-step guide on building a fully automated process to regularly update a BigQuery table — Context Lately, the Roquette Data & Advanced Analytics team has been investigating how Analytical data warehouses such as BigQuery or Snowflake could improve the access to data of our end-users. Such data warehouses offer a global platform to easily ingest, process, and serve structured data to business users or solutions.

In a real-life example, let’s explore how this simple but powerful calculation can help you filter data and, especially, different production batches — General use of Gradient in Machine Learning For sure, if you have been into Data Science for a while… or even for only a few weeks, you have probably heard about how gradient descent and how it can help you find the best coefficients of a linear function or a neural network.

Looking at some handy optimization functions to switch from predictions to prescriptions! — Having a robust model that replicates the ground truth is great, but being able to find the features’ values that would minimize or maximize its response might be even better. Understand why with chocolate chip cookies 🍪! Context Imagine that you have just been hired by a chocolate chip cookies factory! A…

Some models used by R&D, inherited from fundamental theories, might be hard to manipulate or transfer. This real-life example (with code) explains why surrogate models offer an interesting alternative to heavy and complex ones. — Theoric vs Surrogate approaches First, it is important to distinguish the main differences between the two types of models we will be discussing in this article: Theoric Models should be considered as the “phenomenological” ones, inherited from scientific knowledge and expressed as equations, like the (very simple!) one below:

Everything you need to know regarding this new type of challenge, focusing on Data Quality rather than models’ performance — Give honor where honor is due: Andrew Ng If you already know Andrew Ng, you can skip this part! But if you don’t, Andrew Ng is the founder of DeepLearning.AI, Chairman and Co-Founder of Coursera, and a Professor at Stanford University. He has changed the lives of many aspiring data scientists (including me!) when he released his famous…