https://github.com/dennyglee/viz-word-emb

Visualize your word embeddings
https://github.com/dennyglee/viz-word-emb

Last synced: about 2 months ago
JSON representation

Visualize your word embeddings

• Host: GitHub
• URL: https://github.com/dennyglee/viz-word-emb
• Owner: dennyglee
• License: apache-2.0
• Created: 2023-12-04T07:18:25.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-12-04T07:40:04.000Z (about 1 year ago)
• Last Synced: 2024-11-29T06:42:32.090Z (about 2 months ago)
• Language: Python
• Size: 63.5 MB
• Stars: 1
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Visualize your word embeddings (viz-word-emb)

This repository contain various scripts, references, and datasets to help visualize your word embeddings.  These scripts are inspired by the [Generative AI exists because of the transformer](https://ig.ft.com/generative-ai/).

* glove

  * `glove.6B.50d.word2vec.txt`: The original Glove 6B 50d dataset can be downloaded from [GloVe: Global Vectors for Word Representation](https://nlp.stanford.edu/projects/glove/).  This is a copy of the converted `glove.6B.50d.txt` to `word2vec` format to make it easier to work with `gensim`.

* images

  * Contains word embedding visualizations in bar graph format

* scripts

  * `glove_to_word2vec.py`: Converts Glove format to word2vec format; script is based on [glove2word2vec page](https://radimrehurek.com/gensim/scripts/glove2word2vec.html.

  * `word_to_embedding_bar_graph.py`: Finds word embedding of a given word from Glove 6B 50d, scales the values, and plots it as a  1 row 50D (cell) bar graph

    *  Usage: `python3 word_to_embedding_bar_graph.py cycling`