search

declare-lab / multimodal-deep-learning

This repository contains various models targetting multimodal representation learning, multimodal fusion for downstream tasks such as multimodal sentiment analysis.
MIT License
748 stars 153 forks source link

How to do the inference on new data set? #5

Open CaffreyR opened 2 years ago

CaffreyR commented 2 years ago

I have trained the Context-Dependent Sentiment Analysis in User Generated Videos , and it went well! But how can I use the model to infer new data! Thanks! @soujanyaporia

soujanyaporia commented 2 years ago

For that, you need to replicate the feature extraction process for every modality. See some sample codes here: https://github.com/soujanyaporia/MUStARD