Explaining black-box models using attribute importance, PDPs, and LIME

In this article we cover explainability for black-box models and show how to use different methods from the Skater framework to provide insights into the inner...

Building a Named Entity Recognition model using a BiLSTM-CRF network

In this blog post we present the Named Entity Recognition problem and show how a BiLSTM-CRF model can be fitted using a freely available annotated corpus...

Accelerating model velocity through Snowflake Java UDF integration

Over the next decade, the companies that will beat competitors will be “model-driven” businesses. These companies often undertake large data science efforts in order to shift...

Fitting Support Vector Machines via Quadratic Programming

In this blog post we take a deep dive into the internals of Support Vector Machines. We derive a Linear SVM classifier, explain its advantages, and...

ML internals: Synthetic Minority Oversampling (SMOTE) Technique

In this article we discuss why fitting models on imbalanced datasets is problematic, and how class imbalance is typically addressed. We present the inner workings of...

Credit Card Fraud Detection using XGBoost, SMOTE, and threshold moving

In this article, we'll discuss the challenge organizations face around fraud detection, how machine learning can be used to identify and spot anomalies that the human...

On-Demand Spark clusters with GPU acceleration

Apache Spark has become the de facto standard for processing large amounts of stationary and streaming data in a distributed fashion. The addition of the MLlib...

Performing Non-Compartmental Analysis with Julia and Pumas AI

When analysing pharmacokinetic data to determine the degree of exposure of a drug and associated pharmacokinetic parameters (e.g., clearance, elimination half-life, maximum observed concentration ([latex]C_{max}[/latex]), time...

The importance of structure, coding style, and refactoring in notebooks

Notebooks are increasingly crucial in the data scientist's toolbox. Although considered relatively new, their history traces back to systems like Mathematica and MATLAB. This form of...