Skip to content

AutoMLAlgorithmConfig

Source code in docs_src/capo_sagemaker/types/auto_ml_algorithm_config.py 
11
12
13
14
15
class AutoMLAlgorithmConfig(TypedDict, closed=True):
    auto_ml_algorithms: NotRequired[
        "capo_sagemaker.types.auto_ml_algorithms.AutoMLAlgorithms"
    ]
    """<p>The selection of algorithms trained on your dataset to generate the model candidates for an Autopilot job.</p> <ul> <li> <p> <b>For the tabular problem type <code>TabularJobConfig</code>:</b> </p> <note> <p>Selected algorithms must belong to the list corresponding to the training mode set in <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_AutoMLJobConfig.html#sagemaker-Type-AutoMLJobConfig-Mode">AutoMLJobConfig.Mode</a> (<code>ENSEMBLING</code> or <code>HYPERPARAMETER_TUNING</code>). Choose a minimum of 1 algorithm.</p> </note> <ul> <li> <p>In <code>ENSEMBLING</code> mode:</p> <ul> <li> <p>"catboost"</p> </li> <li> <p>"extra-trees"</p> </li> <li> <p>"fastai"</p> </li> <li> <p>"lightgbm"</p> </li> <li> <p>"linear-learner"</p> </li> <li> <p>"nn-torch"</p> </li> <li> <p>"randomforest"</p> </li> <li> <p>"xgboost"</p> </li> </ul> </li> <li> <p>In <code>HYPERPARAMETER_TUNING</code> mode:</p> <ul> <li> <p>"linear-learner"</p> </li> <li> <p>"mlp"</p> </li> <li> <p>"xgboost"</p> </li> </ul> </li> </ul> </li> <li> <p> <b>For the time-series forecasting problem type <code>TimeSeriesForecastingJobConfig</code>:</b> </p> <ul> <li> <p>Choose your algorithms from this list.</p> <ul> <li> <p>"cnn-qr"</p> </li> <li> <p>"deepar"</p> </li> <li> <p>"prophet"</p> </li> <li> <p>"arima"</p> </li> <li> <p>"npts"</p> </li> <li> <p>"ets"</p> </li> </ul> </li> </ul> </li> </ul>"""

auto_ml_algorithms instance-attribute 

auto_ml_algorithms: NotRequired[AutoMLAlgorithms]

The selection of algorithms trained on your dataset to generate the model candidates for an Autopilot job.

  • For the tabular problem type TabularJobConfig:

    Selected algorithms must belong to the list corresponding to the training mode set in AutoMLJobConfig.Mode (ENSEMBLING or HYPERPARAMETER_TUNING). Choose a minimum of 1 algorithm.

    • In ENSEMBLING mode:

      • "catboost"

      • "extra-trees"

      • "fastai"

      • "lightgbm"

      • "linear-learner"

      • "nn-torch"

      • "randomforest"

      • "xgboost"

    • In HYPERPARAMETER_TUNING mode:

      • "linear-learner"

      • "mlp"

      • "xgboost"

  • For the time-series forecasting problem type TimeSeriesForecastingJobConfig:

    • Choose your algorithms from this list.

      • "cnn-qr"

      • "deepar"

      • "prophet"

      • "arima"

      • "npts"

      • "ets"