Setup

Installation

In this section, we walk through the process of setting up infinity-workflows on a local computer so that you can start generating data.

Note

This page uses instructions with pip, the recommended installation tool for Python. You may want to install your packages into an isolated environment using tools like conda or venv. Regardless, your environment will need pip and Python 3.8+ available on your system path.

Clone the infinity-workflows repository, which includes Jupyter notebooks and helper modules for generating synthetic data.

git clone https://github.com/toinfinityai/infinity-workflows.git

---

Note

If you are using conda you can follow the instructions below or run the provided install script:

./install_conda_env.sh

Create an isolated python environment using the tool of your choice and enter it. For example, with conda, you can use the following commands:

conda create --name infinity python=3.9 
conda activate infinity

---

From within the repository directory, install the Python dependencies.

pip install -r requirements.txt --force-reinstall

Note that requirements.txt lists infinity-tools, the python package that facilitates the Infinity API interactions. You can read more about infinity-tools in the Infinity Tools Package section.

---

From within repository directory, launch the jupyter notebook environment

./run_notebooks.sh  # This runs: 'PYTHONPATH="$(pwd)" jupyter notebook'

This script simply makes sure the repository’s modules are part of the PYTHONPATH when jupyter notebook is launched.

---

You have completed installation. Please proceed to the Workflows example to generate data.

Upgrades

To upgrade to the latest verion of infinity-workflows, pull the latest code and update your python environment.

git pull

The --force-reinstall argument below guarantees that no stale wheels from cache are used.

pip install -r requirements.txt --force-reinstall