Configuration Reference

All plugin settings live in conf/<env>/azureml.yml. The file is parsed into KedroAzureMLConfig. For dataset configuration in catalog.yml, see the Datasets reference.

Top-level structure

workspace:    # required
compute:      # required
execution:    # optional
schedules:    # optional
jobs:         # optional

---

workspace

Named Azure ML workspace definitions. A __default__ entry is required.

workspace:
  __default__:
    subscription_id: "00000000-0000-0000-0000-000000000000"
    resource_group: "rg-dev"
    name: "aml-dev"
  prod:
    subscription_id: "11111111-1111-1111-1111-111111111111"
    resource_group: "rg-prod"
    name: "aml-prod"

Each workspace entry (WorkspaceConfig) has the following fields:

Field	Type	Required	Description
subscription_id	string	yes	Azure subscription ID
resource_group	string	yes	Azure resource group name
name	string	yes	Azure ML workspace name

Jobs reference a workspace by name via their workspace field. The __default__ is used when no workspace is specified. See Configure multiple workspaces for a walkthrough.

---

compute

Named compute cluster definitions. A __default__ entry is required.

compute:
  __default__:
    cluster_name: "cpu-cluster"
  gpu:
    cluster_name: "gpu-cluster"

Each compute entry (ClusterConfig) has the following fields:

Field	Type	Required	Description
cluster_name	string	yes	Name of the Azure ML compute cluster

Jobs reference a compute entry by name via their compute field.

Tag-based routing

Kedro node tags can route nodes to specific compute clusters. When a node has a tag that matches a named compute entry, that entry is merged with __default__:

compute:
  __default__:
    cluster_name: "cpu-cluster"
  gpu:
    cluster_name: "gpu-cluster"

A node tagged gpu in your Kedro pipeline will run on gpu-cluster. Nodes without a matching tag fall back to __default__. Fields from the tagged entry override __default__ fields.

---

execution

Code packaging and container settings. All fields are optional.

execution:
  environment: "my-env@latest"
  code_directory: "."
  working_directory: /home/kedro

Field	Default	Description
environment	null	Azure ML environment name (e.g. my-env@latest or my-env:3)
code_directory	null	Local directory to upload as a code snapshot; null disables code upload
working_directory	null	Working directory inside the compute container. Set this when your Azure ML environment expects code at a specific path (e.g. /home/kedro). When null, Azure ML uses its default working directory.

The combination of environment and code_directory determines the deployment flow. When code_directory is set (e.g. "."), the plugin uploads a snapshot of your project and runs it inside the environment (code flow). When code_directory is null, the plugin expects the code to already be baked into the Docker image referenced by environment (image flow). See Deploy from CI/CD for guidance on choosing between the two.

---

schedules

Reusable named schedule definitions. Jobs reference them by name.

schedules:
  business_hours:
    cron:
      expression: "0 9 * * 1-5"
      time_zone: "Europe/London"

Each schedule entry has exactly one of cron or recurrence. See Schedule pipelines for end-to-end setup.

cron

Field	Default	Description
expression	required	Cron expression (e.g. "0 2 * * *")
time_zone	"UTC"	IANA time zone name (e.g. "Europe/London")
start_time	null	ISO 8601 start time
end_time	null	ISO 8601 end time

recurrence

Field	Default	Description
frequency	required	Recurrence unit: "minute", "hour", "day", "week", or "month"
interval	required	Number of frequency units between runs
time_zone	"UTC"	IANA time zone name
start_time	null	ISO 8601 start time
end_time	null	ISO 8601 end time
schedule.hours	null	Hours of the day to trigger
schedule.minutes	null	Minutes of the hour to trigger
schedule.week_days	null	Days of the week to trigger (e.g. ["Monday", "Friday"])

---

jobs

Named job definitions. Each job maps a Kedro pipeline to an Azure ML pipeline submission.

jobs:
  training:
    pipeline:
      pipeline_name: "__default__"
      tags: ["training"]
    experiment_name: "training-experiment"
    display_name: "Daily training"
    compute: "gpu"
    workspace: "prod"
    description: "Run the training pipeline on GPU cluster"
    schedule: "business_hours"
    retry:
      max_retries: 3
      timeout: 3600

Field	Default	Description
pipeline	required	Pipeline selection and filter options (see below)
workspace	null	Named workspace entry; falls back to __default__
experiment_name	null	Azure ML experiment name
display_name	null	Display name shown in Azure ML Studio
compute	null	Named compute entry; falls back to __default__
schedule	null	Inline ScheduleConfig, named schedule string, or null for ad-hoc
retry	null	Retry settings applied to every step (see below)
description	null	Human-readable job description

retry

Optional retry settings applied to every command step in the job. Maps to azure.ai.ml.entities.RetrySettings.

retry:
  max_retries: 3
  timeout: 3600

Field	Default	Description
max_retries	required	Maximum number of retry attempts for failed steps (must be >= 1)
timeout	null	Per-attempt timeout in seconds, or null for no limit

pipeline filter options

These fields correspond to the parameters of Kedro's Pipeline.filter() method.

Field	Default	Description
pipeline_name	"__default__"	Kedro pipeline name
from_nodes	null	Start from these nodes
to_nodes	null	Run up to these nodes
node_names	null	Run only these specific nodes
from_inputs	null	Start from nodes that produce these datasets
to_outputs	null	Run up to nodes that produce these datasets
node_namespaces	null	Filter by namespace
tags	null	Filter by tag

---

Environment variables

The following environment variables are set automatically by the plugin during remote execution. They are reserved and should not be set directly.

Variable	Set by	Description
KEDRO_AZUREML_MLFLOW_ENABLED	Pipeline generator	Set to "1" on each step during remote execution to activate MLflow integration