All you need to know to build a full demo: Hugging Face Hub, Tensorboard, Streamlit, and Hugging Face Spaces — Hello fellow NLP enthusiasts! I’m back at writing articles after a small break, can’t wait to dig into new NLP topics! In this article, we see a complete example of fine-tuning of T5 for generating candidate titles for articles. The model is fine-tuned entirely on Colab, we visualize its training…

BOW, TF-IDF, Word2Vec, Transformer, BERT, GPT, RoBERTa, XLM, Reformer, ELECTRA, T5, and many others — Hello fellow NLP enthusiasts! As the race towards finding better and better neural networks for language modeling continues, I thought it might be a good time to get an overview of the progress made over the years. Enjoy! 😄 Disclaimer: This article is not a complete list of research done…

How to manage the growing complexity of a chatbot project — I recently collaborated on several projects involving chatbots and had the opportunity to discuss with industry experts about the main difficulties that are often encountered in this type of project. …

A text-based and open course with projects, quizzes, and code exercises — Hello, NLP enthusiasts! The previous months have been full of developments for NLPlanet. For those who don’t know about it, NLPlanet is a niche community about Natural Language Processing whose goal is to connect NLP enthusiasts and provide high-quality learning content. …

A really good open-source model for multilanguage speech-to-text — OpenAI recently open-sourced Whisper, a neural network that approaches human-level robustness and accuracy on speech recognition in several languages. The model has been trained on 680,000 hours of multilingual and multitask supervised data collected from the Internet. …

Experimentation, validation, multi-stage deployments, model versioning, performance monitoring, data alerts… — MLOps (Machine Learning Operations) is often a neglected topic in academia and by those taking their first steps in the world of data science, but once you start making your first projects in a company it immediately becomes a hot topic. I recently stumbled upon the paper “Operationalizing Machine Learning…

Semantic search, sentiment analysis, recommender systems, question answering, knowledge graphs, etc. — Let’s go into detail and see a comprehensive list of tasks that NLP can solve today. Keep in mind that many of them can easily be solved by fine-tuning open-source models for most use cases. Text Classification Text classification is the task of automatically assigning a text document to one or more…

A journey across grammars, expert systems, ontologies, statistical models, neural networks, word embeddings, transformers, etc. — What is NLP Natural Language Processing (NLP) is a subfield of artificial intelligence that is concerned with the interactions between computers and human (natural) languages, in particular how to program computers to process and analyze natural language data. NLP is used to apply computational methods to text in order to extract meaning from…

Tips for making a great data science POC — A great proof-of-concept (a.k.a. POC) can help data scientists communicate their ideas to stakeholders and get buy-in for their projects. This is important because data science projects can be expensive and, without funding, your project may never get off the ground. What is a POC? A POC is a short, contained, and relatively inexpensive…