Quickstart

⛵️ Overview

We recommend using Docker to get started.

Docker is a tool that allows you to run Mage in a containerized environment: you can run Mage on any operating system that supports Docker, including Windows, Mac, and Linux. Using Docker means that you don’t have to worry about installing dependencies or configuring your environment. If you’d like to install Mage without Docker, you can use pip or conda.

If you’re familiar with Docker Compose or plan on adding or extending images (e.g. Postgres) in your project, we recommend starting from the Docker compose template. Otherwise, we recommend Docker run.

🪄 Get Mage

Get started

The following is useful if you plan on customizing your image or adding additional services to a Docker network with Mage. Read more about Docker compose here.

Be sure that Docker and Git are installed and run:

git clone https://github.com/mage-ai/compose-quickstart.git mage-quickstart \
&& cd mage-quickstart \
&& cp dev.env .env && rm dev.env \
&& docker compose up

Once the server is running, open http://localhost:6789 in your browser and explore! We recommend one of our guides to for inspiration.

What’s happening?

We’re cloning our Docker Compose quickstart repo, copying dev.env to .env, which is ignored by git in our configuration (to avoid exposing secrets), then running docker compose up to start a Mage server.

Get started

First, be sure Docker is installed. Next, create a new folder for your project, change directory into that folder, and run the following from your terminal:

docker run -it -p 6789:6789 -v $(pwd):/home/src mageai/mageai /app/run_app.sh mage start [project_name]

Once the server is running, open http://localhost:6789 in your browser and explore! If using any cloud provider use external IP address with the port eg 123.456.78:6789. We recommend one of our guides for inspiration.

What’s happening?

We’re executing docker run, which runs a Docker container from the Mage image. additionally, we’re mounting a volume (the -v command) to persist our project files on our container. This mounts the current folder to /home/src in the mage container. We’re executing our docker image mageai/mageai and running a script to start up the container.

Install Mage

pip install mage-ai

_{If you’re installing on Linux and need help, check out this community provided video walkthrough.}

Create new project and launch tool

mage start [project_name]

Open Mage

Open http://localhost:6789 in your browser.

🏃‍♂️ Run your first pipeline

If you haven’t already, open a browser to http://localhost:6789. From the pipelines page, select example_pipeline and open the notebook view by selecting Edit pipeline from the left side nav.

Select the first block by clicking it and select the “play” icon in the top right to run the block. You’ve just ran your first Mage block & loaded data from a dataset!

Blocks are just Python files that live in your project. You can navigate through your blocks on the left side of the screen, drag and drop them into your pipeline, or create new ones from the pipeline editor.

Blocks are reusable, atomic pieces of code that perform certain actions. Changing one block will change it everywhere it's used, but don't worry, it's easy to detach blocks to separate instances if necessary.

Blocks can be used to perform a variety of actions, from simple data transformations to complex machine learning models. You can even use blocks to run custom code, or to connect to external services.

Within a block, the decorated function (custom, loader, transformer, etc) will be executed. The first argument to the function must be a dataframe or JSON serializable object. The function can return a dataframe or JSON serializable object, or it can return nothing.

You can think of this function like a python main() function, when we run the block, this function will be executed. For example, the following block will return the first 10 rows of the dataframe passed in:

from pandas import DataFrame


@transformer
def transform_df(df: DataFrame, *args, **kwargs) -> DataFrame:
    return df.iloc[:10]

Other code outside the block will be executed, but it will not be returned. You can read more about blocks here.

Do the same for the following cells in the pipeline to transform and export the data. Congrats, you’re now a Mage ninja! 🥷

🧙🏻‍♂️ Install Mage dependencies (optional)

Mage also has the following add-on packages:

Package	Install	Description
all	`mage-ai[all]`	install all add-ons
azure	`mage-ai[azure]`	install Azure related packages
clickhouse	`mage-ai[clickhouse]`	use Clickhouse for data import or export
dbt	`mage-ai[dbt]`	install dbt packages
google-cloud-storage	`mage-ai[google-cloud-storage]`	use Google Cloud Storage for data import or export
hdf5	`mage-ai[hdf5]`	process data in HDF5 file format
mysql	`mage-ai[mysql]`	use MySQL for data import or export
postgres	`mage-ai[postgres]`	use PostgreSQL for data import or export
redshift	`mage-ai[redshift]`	use Redshift for data import or export
s3	`mage-ai[s3]`	use S3 for data import or export
snowflake	`mage-ai[snowflake]`	use Snowflake for data import or export
spark	`mage-ai[spark]`	use Spark (EMR) in your Mage pipeline
streaming	`mage-ai[streaming]`	use Streaming pipelines

To install these, run the following command from the Mage terminal:

pip install "mage-ai[spark]"

or add the following to your requirements.txt file:

mage-ai[spark]

You can access the terminal from the side nav on the right in the pipeline editor page. Read more about installing from requirements.txt here.

🧭 Journey on

Navigate to our tutorials to learn more about Mage and how to build your own pipelines or continue exploring our docs for advanced configuration and deployment options.

If you’re interested in connecting a database in Docker, check out our guide for more information.

Get Started

Concepts

Reference

About

⛵️ Overview

🪄 Get Mage

Get started

What’s happening?

Get started

What’s happening?

🏃‍♂️ Run your first pipeline

🧙🏻‍♂️ Install Mage dependencies (optional)

🧭 Journey on

Get Started

Concepts

Reference

About

​⛵️ Overview

​🪄 Get Mage

Get started

What’s happening?

Get started

What’s happening?

​🏃‍♂️ Run your first pipeline

​🧙🏻‍♂️ Install Mage dependencies (optional)

​🧭 Journey on

⛵️ Overview

🪄 Get Mage

🏃‍♂️ Run your first pipeline

🧙🏻‍♂️ Install Mage dependencies (optional)

🧭 Journey on