prefect-learning

This repo has basic examples of prefect

Install prefect

---

Create a separate environment if you prefer. You can use miniconda, pipenv or virtualenv

To install prefect run

pip install prefect

to install it via conda or pienv refer the prefect doc

Running local prefct server & dashboard GUI

---

before starting the server, first run

prefect backend server

This will configure prefect for local orchestration.

To start the server, run

prefect server start

It will start the local server with UI. You can check the dashboard at localhost:8080

You can also specify the port number by giving the option --ui-port port_number

example: prefect server start --ui-port 8085

Running a prefect agent

---

At least one prefect agent is required if you want to run a flow from server.

To run a local agent, run

prefect agent local start

Before running flow, you need to create project (from UI) so that you can register a flow to a project.

Once a project is created, you can register a flow with the server by calling flow.register() (inside a your program/code)