Create a Policy
databricks
, provide a brief description, and then click Next.noarch
or linux-64
architectures only, then click Next.Create a channel for Anaconda's packages
databricks-anaconda
.Create a mirror for Anaconda's packages
https://repo.anaconda.com/pkgs/main
APPLY CHANNEL POLICY
.
How do I know when my mirror is complete?
Create a channel for conda-forge packages
databricks-conda-forge
.Create a mirror for conda-forge packages
https://conda.anaconda.org/conda-forge
APPLY CHANNEL POLICY
.Generate a token for your channels
databricks-token
.databricks-anaconda
channel.databricks-conda-forge
channel.Build a Custom Docker Image
dcs-conda
by running the following command:
dcs-conda
directory and create a Dockerfile
file inside the dcs-conda
directory:
Dockerfile
file, depending on your Databricks Runtime version:
env.yml
file inside the dcs-conda
directory:
env.yml
file:
Launch a Cluster using Databricks Container Service
Enabling Databricks Container Service
enableDcs
to true
, as in the following example:Databricks Runtime for Machine Learning
does not support Databricks Container Service.spark.databricks.unityCatalog.volumes.enabled true
.Docker Image URL examples
<organization>/<repository>:<tag>
<aws-account-id>.dkr.ecr.<region>.amazonaws.com/<repository>:<tag>
<your-registry-name>.azurecr.io/<repository-name>:<tag>
Create a Notebook and connect it to your cluster
Verify your conda installation
!conda --help
runs the command in the current shell. %sh conda --help
starts a subshell, which is useful for multi-line scripts, but might not have the same environment or path.Install MLflow from your Anaconda organization channel