Word Embeddings CLI

This Command Line Interface (CLI) provides functionality to work with word embeddings, specifically utilizing a Word2Vec model for computing similarities between phrases.

Installation

To use this CLI, ensure you have Python 3 installed on your system along with the necessary dependencies. You can install the dependencies by running:

pip install -r requirements.txt

Usage

Calculating Similarities

To calculate similarities between phrases and save the results to a file, use the following command:

python main.py --phrases-file <phrases_file_path> --embeddings-file <user_input> --user_input 

• <phrases_file_path>: Path to the file containing phrases.
• <embeddings_file_path>: Path to the binary file containing the Word2Vec model.
• <user_input>: Phrase to find the closest match for.

Note

• Ensure that the phrases file (phrases.csv in this example) is in the correct format and encoding. Similarly, the embeddings file (word2vec.bin in this example) should be a valid binary file containing the Word2Vec model.
• The output file will be created in the 'data/similarities.csv' destination