Skip to content

README.md

Latest commit

 

History

History
26 lines (20 loc) · 630 Bytes

README.md

File metadata and controls

26 lines (20 loc) · 630 Bytes

Example of embeddings using ELMo on TensorFlow Hub

This repository is an example of word embeddings using ELMo (Embeddings from Language Models) on TensorFlow Hub.

The comparison_test module can calculate cosine similarity between given two sentences.

Usage

  • In case of using exmaple.sh
sh example.sh
  • In case of executing on your terminal
python comparison_test.py [sentence1] [sentence2]

Example

bash-3.2$ python comparison_test.py "people read the book" "the book people read"

[Cosine Similarity]
"people read the book" vs "the book people read"
ELMo: 0.83875865
NNLM: 1.0