Natural Language Processing Course for Thesis/Dissertation Guidance: N
Learn next-generation NLP with transformers for sentiment analysis, Question Answering, similarity search, NER, and much more
What will you learn?
-
Industry-standard NLP using transformer models
-
Build batch question-answering transformer models
-
Perform sentiment analysis with transformer models in PyTorch and TensorFlow
-
Advanced search technologies such as Elasticsearch and Facebook AI Similarity Search (FAISS)
-
Create customized transformer models for specific use cases
-
Measure language model performance using advanced metrics such as ROUGE
-
Vector creation techniques such as BM25 or Dense Passage Retrieval (DPR)
-
Overview of the latest developments in NLP
-
Understand attention and other key transformer components
-
Learn about major transformer models such as BERT
-
Preprocess text data for NLP
-
Named Entity Recognition (NER) using spaCy and transformers
-
Fine-tune language classification models
Requirements
-
Knowledge of Python
-
Experience in data science is a plus
-
Experience in NLP is a plus
Description
Transformer models are the de-facto standard in modern NLP. They have proven themselves to be the most expressive and powerful language models by a large margin, beating all major language-based benchmarks repeatedly.
In this course, we learn everything you need to know to start building state-of-the-art NLP applications using transformer models such as Google AI's BERT, or Facebook AI's DPR.
We cover several major NLP frameworks including:
-
HuggingFace Transformers
-
TensorFlow 2
-
PyTorch
-
SpaCy
-
NLTK
-
Gensim
And learn how to apply transformers to some of the most popular NLP use cases:
-
Language classification/sentiment analysis
-
Named Entity Recognition (NER)



