Installing development tools ​

Gather your credentials ​

After your administrator has onboarded you to Union Cloud, you should have the following at hand:

  • Your Union Cloud credentials.
  • The credentials to access the AWS or GCP account hosting your Union Cloud instance.
  • The URL of your Union Cloud instance. We will refer to this as <union-cloud-host-url> below.

Install Python and Docker ​

Next, make sure that you have the following installed on your local machine:

  • Python: Versions 3.8.x - 3.10.x are supported. Version 3.10.x is used in this guide and is recommended.
  • Conda: In this guide, we use the conda tool (installed via Miniconda) to manage Python versions and virtual environments. You can also use other tools such as pyenv and venv. Using some type of Python virtual environment manager is highly recommended.
  • Docker: Any OCI-compatible container engine like Podman, LXD, or Containerd should also work. Ensure that the associated client daemon is running.

Install Uctl ​

Install the Union Cloud command-line tool, uctl:

To use Homebrew, do this:

shell
brew tap unionai/homebrew-tap
brew install uctl
brew tap unionai/homebrew-tap
brew install uctl

To use curl, set BINDIR to the install location (it defaults to ./bin) and do this:

shell
$ curl -sL https://raw.githubusercontent.com/unionai/uctl/main/install.sh | bash
$ curl -sL https://raw.githubusercontent.com/unionai/uctl/main/install.sh | bash

To download manually, see the UCTL releases.

Note

uctlis an enhanced version of flytectl, the Flyte command-line tool. It adds Union Cloud-specific functionality, letting you manage not only Flyte entities (projects, domains, workflows, tasks, and launch plans) but also Union Cloud-specific entities like users, roles, and Union Cloud configurations.

Set up a Python virtual environment ​

Create a Python virtual environment for our wine-classification example and switch to it:

shell
[~]:base
$ conda create -n wine-classification python=3.10 -y

[~]:base
$ conda activate wine-classification

[~]:wine-classification
$
[~]:base
$ conda create -n wine-classification python=3.10 -y

[~]:base
$ conda activate wine-classification

[~]:wine-classification
$

Note

We useconda to manage the Python version and virtual environments. You are free to use other tools such aspyenv andvenv.

Install Flytekit ​

Finally, install flytekit, Flyte’s Python SDK, within the virtual environment that you just set up:

shell
[~]:wine-classification
$ pip install -U flytekit
[~]:wine-classification
$ pip install -U flytekit