jrbalderrama / pow

This is a MIRROR depository

Privacy-enhancing technologies Hands-on Workshop

Install

This project runs on a Jupyter server compatible with Python 3.7. If you run the project on colab, the following installation is optional (unless you would like to run on colab using local resources).

In order to run the notebooks without colab, first, you might create a virtual environment (optional). Second, install the dependencies, to do so install poetry a packaging and dependency management tool. Third, clone the project sources. Finally, run the poetry install command to setup the required dependencies. The complete installation sequence is as follows (set the PATH accordingly):

# clone the project and enter to the repo dir
git clone https://gitlab.inria.fr/spicy/pow.git pow && cd $_
# create a virtual environment  
python3.8 -m venv ${PATH}/venv-pow
# activate the environment
source ${PATH}/venv-pow/bin/activate 
# install the dependency management tool
pip install poetry
# install dependencies on the environment
poetry install

There is also an conda configuration file environment.yml provided for cases where the management with poetry is not possible, in that case instead of the standard install execute:

# clone the project and enter to the repo dir
git clone https://gitlab.inria.fr/spicy/pow.git pow && cd $_
# create a conda environment 
conda env create -f environment.yml
# activate the environment
conda activate pow

Run

The project is organized in self-contained Jupyter notebooks. These are located in the notebooks directory of the project. There are two ways to run them:

• On colab, click on the links of each notebook to copy the notebook to colab. You require your own Google credentials to execute it. Do not forget to restart the notebook after the installation of the dependencies (pow-api).

• On a local instance of a Jupyter server, after the dependencies installation, launch Jupyter (copy the link displayed in the terminal on a Web browser) and click on a notebook from the list.

  jupyter notebook

In fact, there is a hybrid way to execute the notebooks. On colab you can choose "Connect to local runtime" from the Connect toolbar. To ensure the connection to a local Jupyter server, you have to activate the appropiate extension:

jupyter serverextension enable --py jupyter_http_over_ws