Every major project work organized
All projects are run with Jupyter with Python3 on Ubuntu 16.04.
One needs to have these dependencies:
For Image Conversion from Wav files into PNG (only applicable for subvocalization project), one needs to install SOX
sudo apt-get install sox
- Numpy (pip install numpy)
- Pandas (pip install pandas)
- sklearn (pip install sklearn)
- glob (pip install glob)
- seaborn (pip install seaborn)
- cv2 (pip install cv2)
- matplotlib.pyplot (pip install matplotlib)
- tensorflow (pip install tensorflow)
- PIL (pip install Pillow)
- skimage (pip install scikit-image)
On the terminal, you can type "sudo -i" to be in your sudo mode. Then you can run all those "pip instsall ..." commands. If not, then you can run those commands by typing "sudo pip install ...". IF YOU DON'T HAVE SUDO, IT MAY NOT WORK.
Lastly, for our ML piscine Final Project, we did Subvocalization Classification using CNN. This is the tensorboard graph for my test. I had 97% accuracy on the test.
Enjoy :)