This repository contains code for Universal Sentence Encoder blog post.
First install the dependencies.
pip3 install --quiet "tensorflow>=1.7"
pip3 install --quiet tensorflow-hub
pip3 install --quiet seaborn
To run the message encoder example, use:
python3 embedMessages.py
To run the Semantic Similarity Analysis on Avengers:Infinity Warcast, use:
cd Avengers-Similarity-Analysis
python3 process-script.py
This will process the raw script and remove all the text contained in brackets.
Next, we have to extract the dialogues of the characters.
python3 get-character-lines.py
Finally, use the Universal Sentence Encoder for running Semantic Similarity Analysis.
python3 universal-sentence-encoder.py
This will display and save the Similarity Matrix as Avenger-semantic-similarity.png.
Want to become an expert in AI? AI Courses by OpenCV is a great place to start.
