Retraining churn models presents unique challenges that need special attention Photo by CrowN on Unsplash Retraining machine learning models, especially those focused on customer churn prediction, is an essential step to ensure their relevancy and accuracy over time. However, retraining churn models presents unique challenges that need special attention. Among the most notable is distinguishing between causal effects of interventions — identify customers who stayed due to the proactive retention program to target them exclusively. Intervention Impact on Churn Retraining Consider the follo ..read more

34% Faster Integer to String Conversion Algorithm Are we printing integers fast enough? 1. Introduction In computer programming, converting given integer to a string is a common operation, which should be done for example before printing the integer to the screen, or printing it to any kind of textual file, such as *.xml, *.json, *.csv, *.txt, etc… It is well known that integers (as well as everything else) are stored in computer memory in binary format — as sequences of 0s and 1s. For example: number 12 is represented in memory as “1100”, and number 29 is represented as&n ..read more

In this article, I explore the public transport systems of four selected cities relying on General Transit Feed Specification and various tools of spatial data science. I picked four cities in this notebook, Budapest, Berlin, Stockholm, and Toronto, to overview their public transport system using publicly available GTFS (General Transit Feed Specification) data. This notebook aims to serve as an introductory tutorial on accessing, manipulating, aggregating, and visualising public transport data using Pandas, GeoPandas, and other standard data science tools to derive insights about public ..read more

Little Heavy | Image by AuthorLeveraging the power of macOS GPUs with the Metal.jl Framework. Introduction Just last year, we were introduced to the Metal.jl Framework, a GPU backend for Apple Hardware. This is exciting news for Julia practitioners looking to leverage the full potential of their macOS M-series chips. In particular, data scientists and ML engineers can speed up their computational workflows by tapping into the parallel processing power of GPUs, resulting in faster training and inference times. The integration of Metal.jl into the Julia ecosystem signifies an important push ..read more

If you are working in another field, how can you transition to data analytics? You may have a university degree in an unrelated field, or have been working in a completely different domain. You may be interested in transitioning into a data analysis role for a while, but do not know where to start. If this sounds like you, keep on reading! ? Photo by Myriam Jessier on UnsplashTwo ways to get into data analytics Essentially there are 2 ways to get into data analytics: (1) Completely self-taught: then cleverly combine analytics skills with your current domain knowledge either from you ..read more

With A Tail of Cat Food Preferences Photo by Anastasiia Rozumna on Unsplash Welcome to the ‘Courage to learn ML’. This series aims to simplify complex machine learning concepts, presenting them as a relaxed and informative dialogue, much like the engaging style of “The Courage to Be Disliked,” but with a focus on ML. In this installment of our series, our mentor-learner duo dives into a fresh discussion on statistical concepts like MLE and MAP. This discussion will lay the groundwork for us to gain a new perspective on our previous exploration of L1 & L2 Regularization. For a com ..read more

Photo by ThisisEngineering RAEng on Unsplash Describing the nature with the help of analytical expressions verified through experiments has been a hallmark of the success of science especially in physics from fundamental law of gravitation to quantum mechanics and beyond. As challenges such as climate change, fusion, and computational biology pivot our focus toward more compute, there is a growing need for concise yet robust reduced models that maintain physical consistency at a lower cost. Scientific machine learning is an emergent field which promises to provide such solutions. This art ..read more

Graph neural networks (GNNs) and large language models (LLMs) have emerged as two major branches of artificial intelligence, achieving immense success in learning from graph-structured and natural language data respectively. As graph-structured and natural language data become increasingly interconnected in real-world applications, there is a growing need for artificial intelligence systems that can perform multi-modal reasoning. This article explores integrated graph-language architectures that combine the complementary strengths of graph neural networks (GNNs) and large language models (LLMs ..read more

The definitive guide for beginners Photo by Sebastian Svenson on Unsplash Data modelling is an essential part of data engineering. In this story, I would like to talk about different data models, the role of SQL in data transformation and the data enrichment process. SQL is a powerful tool that helps to manipulate data. With data transformation pipelines we can transform and enrich data loaded into our data platform. We will discuss various methods of data manipulation, scheduling and incremental table updates. In order to make this process efficient we would want to know a few essential ..read more

Good engineers, bad engineers, and evil engineers — an anecdote for data leaders My golden framework to differentiate the good, the bad, and the evil engineers in all fields, including data Image by author (generated using Canva’s Magic Media app) To engineer is to design or build something using scientific principles— Cambridge Dictionary. We all love good engineers, they build fantastic bridges, roads, rockets, applications, and data structures that make our lives easier and enjoyable every day. By the same logic, bad engineers will not make lives much better. If we hire them ..read more