Coursera NLP specialization Learning Notes
This repository contains personal notes about the NLP (Natural Language Processing) Specialization of DeepLearning.AI.
The purpose of the course is to lay the foundations to understand seq2seq models and attention models. The course is divided in 4+1 main courses.
-
The first course uses Logistic Regression and Bayes classifiers to perform sentiment classification. This course also deep dives about the different word representation techniques. Finally a basic machine learning translation system is built frond the ground and the first search method, sensitive hashing, is introduced.
-
The second course focus about probability theory in NLP models. We explore the different algorithms that can be used to predict the next word in a sentence as well as how to compare these predictions against others.
-
The third course is centered around seq2seq models.
-
The final course teaches the attention models and the attention mechanism. This course also gives us the background to implement the most common NLP use cases like text summarization, machine translation, chatbots and question and answering.
-
Course 5 is a practical course where we use Azure AI services to train, efficient fine-tuning, deploy NLP models into the cloud. Besides we explore fundational models, prompt-engineering, retrieval augment generation (RAG)