For machine learning (ML) practitioners, one of the most important and typical tasks is full-text classification (FTC), a technique where you assign a set of categories or tags to an entire row of text. The way you do this apparently simple task has a direct impact on how we build... Read more

At ODSC London 2018, Dr. Michael Swarbrick Jones, Ph.D. gave a technical lecture on the latest in large-scale multilabel classification and how practitioners should be managing their hyperparameters and data to get the most out of their models, focusing specifically on problems where there are thousands or tens of thousands... Read more

TLDR AUC is a good starting metric when comparing the performance of two models but it does not always tell the whole story NRI looks at the new models ability to correctly reclassify cancers and benigns and should be used alongside AUC IDI quantifies improvement of the slopes of the discrimination curves and... Read more

Deep learning is everywhere. From sales forecasting to segmenting skin diseases on image data — there’s nothing deep learning algorithms can’t do, given quality data. If deep learning and TensorFlow are new to you, you’re in the right place. This article will show you the entire process of building a classification model... Read more

TLDR AUC is a good starting metric when comparing the performance of two models but it does not always tell the whole story NRI looks at the new models ability to correctly reclassify cancers and benigns and should be used alongside AUC IDI quantifies improvement of the slopes of the discrimination curves and... Read more

In this post, I’ll explain how to convert time-series signals into spectrograms and scaleograms, which are image representations of those signals that contain both frequency and time information. In a future post, we’ll use the images created here to classify the signals. I’ll explain the intuition and math used to... Read more

This article discusses AdaptNLP. In Natural Language Processing (NLP), language models such as ULMFiT, BERT, and GPT have become the foundation of many solutions for common NLP tasks. The benefit of language models is their ability to be pre-trained with a general understanding of language, such that users can fine-tune... Read more

This article is the last of a series in which I cover the whole process of developing a machine learning project. If you have not read the previous two articles, I strongly encourage you to do it here and here. The project involves the creation of a real-time web application that gathers data from several newspapers and... Read more

Paolo is a speaker for ODSC East 2020 this April 13-17. Be sure to check out his talk, “Guided Labeling: Human-in-the-Loop Label Generation with Active Learning and Weak Supervision,” at this upcoming event! One of the key challenges of utilizing supervised machine learning for real-world use cases is that most... Read more

This article is the first of a series in which I will cover the whole process of developing a machine learning project. This one focuses on training a supervised learning text classification model in Python. The motivation behind writing these articles is the following: as a learning data scientist who has been... Read more