AutoML for Natural Language Processing with Abhishek Thakur - #475
Published:Apr 15, 2021 16:44
•1 min read
•Practical AI
Analysis
This article summarizes a podcast episode featuring Abhishek Thakur, a machine learning engineer at Hugging Face and a Kaggle Grandmaster. The discussion covers Thakur's journey in Kaggle competitions, his transition to a full-time practitioner, and his current work on AutoNLP at Hugging Face. The episode explores the goals, problem domain, and performance of AutoNLP compared to hand-crafted models. It also mentions Thakur's book, "Approaching (Almost) Any Machine Learning Problem." The article provides a concise overview of the podcast's key topics, highlighting the intersection of competitive machine learning, practical application, and the development of automated NLP tools.
Key Takeaways
- •The podcast episode features Abhishek Thakur, a Kaggle Grandmaster and machine learning engineer at Hugging Face.
- •The discussion covers Thakur's experience in Kaggle competitions and his transition to a full-time role.
- •A significant portion of the episode focuses on AutoNLP, its goals, and its performance compared to hand-crafted models.
Reference
“We talk through the goals of the project, the primary problem domain, and how the results of AutoNLP compare with those from hand-crafted models.”