NLP to LLM from Basic Principles: NLPlanet NLP and ML pt.4
Preface:
I am undertaking a Natural Language Processing and Machine Learning course that uses NLTK and HuggingFace. I have taken notes on their lessons and would like to share them. Although I tried to be in-depth and transformative with my notebook, I strongly encourage following their course as it is detailed and well-formatted.
Credit:
Course Link:
Welcome to the Course — Practical NLP with Python (nlplanet.org)
(Section link provided further down)
Author:
Note Information:
This is a Jupyter Notebook uploaded to my Github as provided in the link further down.
Notes include definitions, examples, and explanations.
Uses Sentence Transformers, Hugging Face Hub, Pandas, NumPy, Scikit-learn, and Plotly.
Title:
Section 1.12
Representing Text as Vectors with Word Embeddings
Section Link:
1.12 Representing Texts as Vectors: Word Embeddings — Practical NLP with Python (nlplanet.org)
Information Covered (All Defined in Notes):
- Word Embeddings
- Word Embedding models
- Context-independent Embedding
- Context-dependent Embedding
- Pre-trained Models
- Finetuning
- Sentence Embeddings
- Sentence Transformers
- Cosine Similarity
- MTEB
- Analyzing Similarity
- Plotly Visualizations
- Datasets
- DataFrames
- Bag of Words vs Embeddings comparison