cli-automl-forecasting-task-energy-demand.yml

$schema: https://azuremlsdk2.blob.core.windows.net/preview/0.0.1/autoMLJob.schema.json
type: automl

experiment_name: dpv2-cli-automl-forecasting-experiment
# name: dpv2-sdk-forecasting-train-job-01
description: A Time-Series Forecasting job using energy demand dataset

compute: azureml:cpu-cluster

task: forecasting
primary_metric: r2_score

log_verbosity: debug

target_column_name: "demand"
training_data:
  path: "./training-mltable-folder"
  type: mltable
validation_data:
  path: "./validation-mltable-folder"
  type: mltable

limits:
  timeout_minutes: 180
  max_trials: 10
  max_concurrent_trials: 5
  trial_timeout_minutes: 20
  enable_early_termination: true

forecasting:
  time_column_name: "timeStamp"
  forecast_horizon: 48
  frequency: "H"
  target_lags: [12]
  target_rolling_window_size: 4
  # TO TRY: ADDITIONAL FORECASTING TRAINING PARAMS ---
  # time_series_id_column_names=["tid1", "tid2", "tid2"],
  # short_series_handling_config=ShortSeriesHandlingConfiguration.DROP,
  # use_stl="season",
  # seasonality=3,

training:
  enable_model_explainability: true
  enable_stack_ensemble: false
  allowed_training_algorithms:
    - xg_boost_regressor