kshitijdshah99
commented
7 months ago Hey, @hubertooi I would love to contribute to this issue. I have had a previous experience in building a Transformer architecture from scratch hence looking forward to work with you in this issue. Can you tell me what exactly am I expected to do?
Video Title: Natural Language Processing: Trends, Challenges and Opportunities | PyData Global 2021 Link: https://youtu.be/Y2WZEV-Ds-o
0:04 - PyData Co-Chair Welcome Remarks 0:18 - Speaker Marco Bonzanini Intro 1:29 - Speaker Background 2:47 - Natural Language Processing Intro 5:32 - Challenges of Language 6:30 - Annotated Data 9:08 - History and summary of NLP models 11:20 - Evolution of NLP models 11:39 - Bag-of-words 12:03 - Word embeddings 13:22 - ML models evolution 13:45 - RNN/LSTM models 14:30 - Transformers models 15:11 - Transformers architecture 17:49 - 2017 Transformers popularity boom 18:42 - NLP Semi-Supervised Sequence Learning models 19:29 - Can Language Models Be Too Big? 21:13 - Python NLP Ecosystem Libraries 23:15 - Thank You 23:56 - Open Discussion and Q&A Session 24:41 - Q1) How to get started with NLP with simple classifier? 27:06 - Q2) Share your view about NLP on non-english foreign languages? 29:06 - Q3) What books for NLP reference? 29:40 - Book 1: Foundations of Statistical Natural Language Processing 30:40 - Book 2: Deep Learning with Python 31:15 - NLTK.org 32:05 - Q4) What do you use to interpretation of transformers? 33:33 - What do you use NLP for and challenges to implement NLP solution? 40:50 - Q5) Main challenge is capturing domain specific context 46:21 - Use case question example using NLP 51:15 - Q6) Ways to Reduce bias in modeling? 52:21 - Q7) Is there any added value for NLP research? 56:19 - Ending