I try to run my machine-learning code on databricks(community version) and need to use the Orange3 data-mining library. However, when I tried to create the orange3 library, it gives an error like this:
Error: java.lang.RuntimeException: Installation failed with message: Collecting orange3 Downloading Orange3-3.8.0.tar.gz (34.0MB) Complete output from command python setup.py egg_info: Orange requires Python >= 3.4 ---------------------------------------- mesg: ttyname failed: Inappropriate ioctl for device Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-build-DnpIcU/orange3/
I guess that means I need python3 to install the library, but the default version on databricks is python2. I wonder How I can change it to python3.
I also searched on databricks forum and find this post(https://docs.databricks.com/api/latest/examples.html#create-a-python-3-cluster).
However, I still don't quite understand how to create the python3-cluster. Does it mean I need to use Cluster API to customize a cluster, which need to charge extra money?
Appreciate any help!!