Deep Learning Demo
Various projects for demonstrating deep learning technology.
Installation
Install git
# for Windows, please go to the link below to download and install git for Windows
https://git-scm.com/download/win
# For Mac, install git:
git --version
# The above should prompt you to install git.
Install cmake for openCV
For Windows, download the cmake-3.10.3-win64-x64.msi installer and execute the file.
For Mac, click on cmake-3.10.3-Darwin-x86_64.dmg and install like a normal Mac app.
Install Miniconda (Python 3.6)
# For Windows go to https://conda.io/miniconda.html, download and install Python 3.6 installer
# For Mac, download the package
wget https://repo.continuum.io/miniconda/Miniconda3-latest-MacOSX-x86_64.sh
# execute the shell script
bash Miniconda3-latest-MacOSX-x86_64.sh
Keep pressing Enter until it asks you to answer (yes/no), answer yes.
# once you finished installing, remove the downloaded shell script as its not needed anymore.
rm Miniconda3-latest-MacOSX-x86_64.sh
Remember to close the current terminal and reopen again for conda to work!
Git clone the demo repository
# git clone the repo and move into the directory
git clone https://github.com/marcoleewow/deep-learning-demo.git && cd deep-learning-demo
Create virtual environment using Miniconda
conda env create -f environment.yml
Activate the conda environment
# For Windows:
activate myenv
# macOS and Linux:
source activate myenv
You have now successfully set up everything!
Try out the demo!
For face recognition:
python face-recognition/facerec_from_webcam_faster.py
For blurring face:
python face-recognition/blur_faces_on_webcam.py
For multi style transfer (please remember to download the model first), note that this demo is quite slow on CPU:
# download the model
bash style-transfer/models/download_model.sh
# run the script
python style-transfer/camera_demo.py
For chatbot:
python chatbot/chatter_bot.py
Finish running demo
# deactivate your virtualenv
source deactivate