Rosette API Text Embeddings Sample Code
This is a little python code to show how to calculate the similarity between words by computing the cosine similarity (using numpy) between the words' embeddings, returned from the Rosette API's new
/text-embedding endpoint. The call to the API uses the 1.3 version of the python binding, so be sure to install that package via
$ pip install rosette-api or
--upgrade via pip to get the latest.
To try it out
- Clone the repo and open the files in your favorite text editor/python IDE.
cosine_similarity.py, replace the
[your key here]with your Rosette API key and save.
test_embeddings.pyvia your python IDE or command line:
$ python test_embeddings.py
Customize for your data
test_embeddings.py to compare words OR longer text you might be interested in to see how their embeddings compare. And if you find anything interesting, let us know! Find us at support.rosette.com or email@example.com.