Our public demo instance is the quickest way to see LynxKite in action:
This instance showcases a number of prepared wizards and workspaces. If you want to try creating your own workspaces or using the Python API, start your own demo instance.
This option starts up your personal LynxKite instance in the cloud.
Go to our download page.
Click the “My demo instance” button, log in with a Google Account.
Wait for your instance to start up. This can take a minute or two — we are starting a brand new VM just for you!
You can either open the LynxKite GUI, or a JupyterLab interface where you can check out some notebooks to see LynxKite’s Python API.
If something doesn’t work, let us know at lynxkite@lynxkite.com!
Make sure you have Docker installed. (Supports Windows, macOS, and Linux.)
Start the LynxKite Docker container:
docker run \
-p 2200:2200 \
-v ~/lynxkite_meta:/metadata -v ~/lynxkite_data:/data \
-e KITE_MASTER_MEMORY_MB=1024 \
--name lynxkite lynxkite/lynxkite
LynxKite stores metadata (such as your workspaces) and data (such as graphs) locally
in the default configuration. Feel free to change lynxkite_meta and lynxkite_data in the example
to store them under a different path.
The KITE_MASTER_MEMORY_MB environment variable controls the amount of memory used by LynxKite.
The default setting is 1 GB.
After a few seconds you can navigate to http://localhost:2200/ and see the LynxKite UI.
If you cannot get it to work, let us know at lynxkite@lynxkite.com! If it does work, try LynxKite Tutorial: Flight Routes.
Make sure Java 8 is installed.
Make sure Apache Spark 3.3.2 is installed. (pip install pyspark==3.3.2 or download)
Download LynxKite from the download page using the “Download for Linux” button.
Set environment variables to control where LynxKite stores its data and how much memory it uses.
export KITE_META_DIR=$HOME/lynxkite/meta
export KITE_DATA_DIR=file:$HOME/lynxkite/data/
export KITE_MASTER_MEMORY_MB=4000
The above values are just some sensible defaults. You can look up what they and various other settings mean in the Admin Manual.
Run LynxKite:
spark-submit lynxkite-5.4.1.jar
After a few seconds you can navigate to http://localhost:2200/ and see the LynxKite UI.
If you want to use advanced machine learning boxes in LynxKite, install our Python dependencies, PyArrow and PyTorch Geometric.
If you cannot get it to work, let us know at lynxkite@lynxkite.com! If it does work, try LynxKite Tutorial: Flight Routes.