![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhc7hJ9vp5J3aSxR-iLVqJimGEIftBsuRE9Mzwph6YYdPQaR6YbisbfjRLUSgeCKsxqRaWJfI-4w6DbEYgMPWB7kWB0hUzew8MkbWN_bscM6QuHly9EZkjq19ju_0v2i7onVI66DEdFY1bV/s400/100-copy+figure.png)
Open Google Colaboratory (Colab) in a separate tab in your browser.
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjtNnSyr90LSUYXTO-s_aQAsPRFcLchhTsMlvR_0QdQsFmc7QKk8_GRaTnZGm1Q7StUDe-jvi5VE1A8X-x8O2EOJlVQ8dePmqZLbE6G6mwVp7WA7y9NIeqCg1dMkBdtgYxdCfx5WNh74CeD/s400/110-colab.png)
Execute the first cell to mount your Google drive.
from google.colab import drive
drive.mount('/content/gdrive')
Skip the second cell.
Run the third cell to move into the Expression-Net directory.
%cd "/content/gdrive/My Drive/Expression-Net"
To analyze the example images, we can either run the fourth cell
!python autoface.py ./input.csv
Alternatively, we can run the fifth cell
!python autoface.py ./figures/donald_trump.csv
!python autoface.py ./figures/hillary_clinton.csv
The autoface.py script creates a json file in the output_params folder. Switch to the Google Drive tab and step into this folder. Download the json files and put them in the directory AUTOFACE_DIR/parameters. By default AUTOFACE_DIR is the Autoface folder in your Documents directory.
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhAHZZ0HqkbxLApkEqQ3pb5TleWN9JWXUw4mbS8NmVqrQIfpr3D70HF5cx7LxIzOaI3VR9wASY4-QPgPgUrkB0647L7P86NMwq74sTdYmWYffKc-vkUrPbaSoiLtfQLfrFvSMLpZ6WLZB6J/s400/120-output-params.png)
![](https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEg7r2mihm8aziyWaOz4VB9m9zyRBdWRaP-cudqabRAQ-j1fNbnjwxpaJ2utiaqfYHVlZJl3JB2SDsoYVtIYqCONbo9tSST57RmFr3_PxBV-STasS98DJ9tW6anElgLUuYxPbOV9WBgzDsF0/s400/130-parameters.png)