https://github.com/shilangyu/aif-emotion-detection

Emotion detection project for the Artificial Inteligence Fundamentals course at WUT
https://github.com/shilangyu/aif-emotion-detection

emotion-recognition jupyter-notebook tensorflowjs

Last synced: 29 days ago
JSON representation

Emotion detection project for the Artificial Inteligence Fundamentals course at WUT

• Host: GitHub
• URL: https://github.com/shilangyu/aif-emotion-detection
• Owner: shilangyu
• Created: 2022-03-23T13:15:25.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2024-02-11T21:11:16.000Z (9 months ago)
• Last Synced: 2024-10-06T11:41:03.173Z (about 1 month ago)
• Topics: emotion-recognition, jupyter-notebook, tensorflowjs
• Language: Jupyter Notebook
• Homepage: http://github.shilangyu.dev/AIF-emotion-detection/
• Size: 17.1 MB
• Stars: 2
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Emotion detection

[![](https://github.com/shilangyu/AIF-emotion-detection/workflows/image-demo-cd/badge.svg)](https://github.com/shilangyu/AIF-emotion-detection/actions)

Final rendered report can be found [here](report/report.pdf).

Uses [Pipenv](https://pipenv.pypa.io/en/latest/) for python dependency management.

## Text

More info in [the notebook](./text/text_emotion_detection.ipynb).

In `text/` install dependencies with `pipenv install`, then start the jupyter notebook with `pipenv run jupyter notebook`.

## Images

More info in [the notebook](./image/image_emotion_detection.ipynb) and the [demo README](image/demo).

In `image/` install dependencies with `pipenv install`, then start the jupyter notebook with `pipenv run jupyter notebook`.

The final model can be found [in the releases tab](https://github.com/shilangyu/AIF-emotion-detection/releases):

- `final_model.h5` - all the final weights for the keras model

- `tfjs_model.zip` - same model but converted to tensorflow.js using `pipenv run tensorflowjs_converter --input_format keras final_model.h5 tfjs/`