This repository was archived by the owner on Oct 31, 2023. It is now read-only.

Description
Hi,
I want to use Infersent to measure the similarity scores on the fly - i.e. as and when the user requests similarity scores.
The issue I have with this is the time taken to build the vocab prior to using the model.
Is there any way to save a model after building the vocab? So that it can be loaded and used without the need to continually repeat this step. Or is there a better method to use in this way?
Thanks