davebulaval/nlp-progress

Repository to track the progress in Natural Language Processing (NLP), including the datasets and the current state-of-the-art for the most common NLP tasks.

Hi there, I'm David - aka davebulaval

I’m a Ph.D. student, ML researcher, founder of an AI podcast, and an AI consultant!

• I’m working on my academic research, Poutyne, Deepparse and MeaningBERT!
• I’m developing my skills in NLP
• I’m developing my consulting firm with colleagues: Baseline
• I’m developing a bilingual AI podcast: OpenLayer
• I’m developing a bilingual blog site with colleagues: .Layer

Framework I work on

• Poutyne: A simplified framework for PyTorch and handles much of the boilerplating code needed to train neural networks.
• Deepparse: A state-of-the-art library for parsing multinational street addresses using deep learning.
• notif: A easy-to-use package to send notifications from a Python script.

Other Packages

• Actuarial Symbol: Typesetting actuarial symbols easily and consistently in LaTeX.
• nCopula: Construct and manipulate hierarchical Archimedean copulas with multivariate compound distributions.

Presentations I did

• Reproductibilité en apprentissage automatique
• Configuration and results management with MLflow, Hydra and Poutyne (also in French)
• Levraging Subword Embeddings for Multinational Address Parsing
• Detection of Duplicates Among Non-structured Data From Different Data Sources
• Bonnes pratiques & Git

Articles:

• "FIJO": a French Insurance Soft Skill Detection Dataset
• Multinational Address Parsing: A Zero-Shot Evaluation
• Generating Intelligible Plumitifs Descriptions: Use Case Application with Ethical Considerations
• Leveraging Subword Embeddings for Multinational Address Parsing
• A Robust Self-Learning Method for Fully Unsupervised Cross-Lingual Mappings of Word Embeddings: Making the Method Robustly Reproducible as Well

📺 Latest YouTube Videos

• OpenLayer - #43 | Véronique Tremblay
• OpenLayer - #42 | Gabrielle Vachon
• OpenLayer - #41 | Ilyes Talbi
• OpenLayer - #40 | François Pelletier
• OpenLayer - #39 | Vanessa Deschênes

📌 GitHub Stats