• Introduction to time series forecasting in Python with FB's Prophet

    Introduction to forecasting with FB Prophet Prophet is a forecasting tool developed by Facebook to quickly forecast time series data, available in R and Python. In this post I’ll walk you through a quick example of how to forecast U.S. candy sales using Prophet and Python. First, we’ll read...
  • A Brief History of Machine Learning Models Explainability

    If software ate the world, models will run it. But are we ready to be controlled by blackbox intelligent softwares? Probably not. And this is fair. We, as human, need to understand how AI works — especially when it drives our behaviours or businesses. That’s why in a previous post, we...
  • Kettlebells Muscular Endurance for Combat Athletes

    By Scott Shetler: One of the biggest mistakes I see combat athletes make in their conditioning programs is to overlook the importance of muscular endurance. Muscular endurance is not the same as aerobic capacity. An athlete can utilize exercises like running, cycling, or other forms of cardiovascular exercise to...
  • ECCV 2018 conference recap

    Group Norm Group Norm (GN) is another normalisations scheme, in a way similar to BatchNorm (BN, widely used at Scortex).  BatchNorm has several failure cases:   small batch size (typically for segmentation of large images) when the batch can vary or the notion of batch is ill posed: varying batch...
  • New Politics and Philosophy Podcast: The Badlands

    This is a guest post by Toby Napoletano, a philosophy PhD who is working on a politics and philosophy podcast that is the subject of this post.     I’ve taught quite a few introductory ethics courses to undergraduates over the past five years or so. During that time, I’ve had the good...
  • Chip cards: example of predicting invariables

    A few years ago, I wrote an article for Harvard Business Review, pointing out that data scientists should pay more attention to predicting “invariables” rather than predicting “variables”. See here.  The gist is: it is a lot more reliable to predict an outcome that is invariable, and there are...
  • Where do data scientists come from? Indeed.com research

    Our previous article in this series on Data Science Titles made the case that there’s no such thing as a data scientist, but instead, the phrase “data scientist” has come to represent a number of distinct roles. In addition to different skills and job duties, we’d like to know...
  • Data Notes: How Do Autoencoders Work?

    Autoencoders, convolutions, and predicting Google’s stock price: Enjoy these new, intriguing, and overlooked datasets and kernels   1. 🤖 How Autoencoders Work: Intro and Use Cases (link) 2. 🕋 3D Convolutions: Understanding and Implementation (link) 3. 🕑 Hourly Time Series Forecasting with Prophet (link) 4. 📈 Google Stock Price Prediction – RNN (link)...
  • Rhea: Efficient environment variables management and typing for python

    Efficient environment variables management and typing for python. Installation pip install -U rhea Features Typed retrieval of environment variables. Handling of optional, secret, and local variables. Reading from different sources: os, json files, yaml files. Collection of parsed parameters. Usage Reading typed values from a params from rhea import...