Metadata-Version: 2.1
Name: democlassi
Version: 0.5.1
Summary: Collection of my python functions for training pytorch models to classify emotion, age, race, gender
Home-page: https://github.com/AlkaSaliss/DEmoClassi
Author: A. Alka M. Salissou
Author-email: alkasalissou@hotmail.com
License: LICENSE
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3.6
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
Requires-Dist: torch (==1.8.1)
Requires-Dist: torchvision (==0.9.1)
Requires-Dist: pytorch-ignite (==0.3.0)
Requires-Dist: imutils (==0.5.4)
Requires-Dist: dlib (==19.22.0)
Requires-Dist: kaggle
Requires-Dist: tensorboard
Requires-Dist: tensorboardX
Requires-Dist: tqdm
Requires-Dist: pillow (==8.2.0)
Requires-Dist: matplotlib
Requires-Dist: scikit-learn (==0.24.2)
Requires-Dist: scikit-image (==0.18.1)
Requires-Dist: opencv-python
Requires-Dist: pandas

# DEmoClassi
DEmoClassi stands for **D**emographic (age, gender, race) and **Emo**tions 
(happy, sad, angry, ...) **Classi**fication.

This repository is an attempt to tackle the task of predicting the facial expression,
age gender and race prediction from  facial images.

As I was planning to train the models on Google Colab platform, I decided to gather 
all my utility functions/classes in a package I named democlassi which is pip 
installable via `pip install --upgrade democlassi`.

The Training is done using Pytorch and pytorch-ignite and more 
information about the data used, the training process and live prediction using 
pretrained models and opencv can be found on 
[this page](https://alkasaliss.github.io/DEmoClassi/).


Some sample predictions : 

![png](docs/samples/neutral_shot.png)
![png](docs/samples/happy_shot.png)
![png](docs/samples/angry_shot_new.png)
![png](docs/samples/sad_sht1.png)
![png](docs/samples/surprised_shot.png)


