An introduction to Deep Learning NLP modeling using tensorflow/pytorch.
NLP is fast becoming ubiquitous in the workplace and in everyday life. The goal of this course is to provide an introduction to deep learning for Natural Language Processing (NLP). We use a variety of datasets to explore sentiment analysis, classification, topic modeling, clustering and transfer learning of existing models. We explore core concepts: text vectorization, embedding, transformer and feature extraction applied to different type of networks including fully connected models, LSTM, convolutional layers and BERT. The goal of the course is to demystify concepts and show how to practically build and train NLP deep learning models applied to specific datasets. In the last section, we examine how to leverage LLM in NLP deep learning modeling applications.
- Sentiment analysis modeling: intro to NLP notebook for intro to nlp
- DL NLP intro dense, recurrent and cnn networks notebook for intro to Deep Learning NLP
- Tansfer learning NLP: BERT and Universal Sentence Encoder
- Topic modeling intro: NMF, LSA and LDA
- Topic modeling: BERTopic and Top2Vec notebook for intro to DL topic modeling
- Topic modeling and LLM