I read into causal inference recently and since I didn’t really have a use-case for it right now, I played around with some data and some causal graphs. In this article, I looked at some causal graphs from the “Book of Why” Chapter 4 by Judea Pearl and Dana Mackenzie and created simulated data based of them. Read more

The shapley value is a popular and useful method to explain machine learning models. The shapley value of a feature is the average contribution of a feature value to the prediction. In this article I’ll show you how to compute shapley values from scratch. Read more

In this short article, you will learn, how and when to test the error message of an exception with pytest.

In this article, we look at SimCSE, a simple contrastive sentence embedding framework, which can be used to produce superior sentence embeddings, from either unlabeled or labeled data. The idea behind the unsupervised SimCSE is to simply predicts the input sentence itself, with only dropout used as noise. Read more

Language models gained popularity in NLP in the recent years. Sometimes you might have enough data and want to train a language model like BERT or RoBERTa from scratch. While there are many tutorials about tokenization and on how to train the model, there is not much information about how to load the data into the model. This guide aims to close this gap.

In this article we sample from pre-trained transformers to augment small, labeled text datasets for named entity recognition.

This time, I’m going to talk about how to approach general NLP problems. But we’re not going to look at the standard tips which are tosed around on the internet, for example on platforms like kaggle. Read more

In a recent article, we saw how to implement a basic validation pipeline for text data. Once a machine learning model has been deployed its behavior must be monitored. The predictive performance is expected to degrade over time as the environment changes. Read more

Today, I’m going to talk about topic models in NLP. Specifically we will see how the Latent Dirichlet Allocation model works and we will implement it from scratch in numpy. What is a topic model? Assume we are given a large collections of documents. Read more