create_tuner_stats

airt.keras.experiments.create_tuner_stats(tuner: Tuner, *, num_models: int = 10, max_epochs: int = 50, batch_size: int = 8, patience: int = 10, verbose: int = 0) -> pd.DataFrame

Calculates statistics for the best models found by Keras Tuner

Parameters:

Name	Type	Description	Default
tuner	Tuner	an instance of Keras Tuner	required
num_models	int	number of best models to use for calculating statistics	10
max_epochs	int	maximum number of epochs used in runs	50
batch_size	int	batch_size	8
patience	int	maximum number of epochs with worse objective before stopping trial early	10
verbose	int	verbosity level of Model.fit function	0

Returns:

Type	Description
pd.DataFrame	A dataframe with statistics

Source code in airt/keras/experiments.py

def create_tuner_stats(
    tuner: Tuner,
    *,
    num_models: int = 10,
    max_epochs: int = 50,
    batch_size: int = 8,
    patience: int = 10,
    verbose: int = 0,
) -> pd.DataFrame:
    """Calculates statistics for the best models found by Keras Tuner

    Args:
        tuner: an instance of Keras Tuner
        num_models: number of best models to use for calculating statistics
        max_epochs: maximum number of epochs used in runs
        batch_size: batch_size
        patience: maximum number of epochs with worse objective before stopping trial early
        verbose: verbosity level of `Model.fit` function

    Returns:
        A dataframe with statistics
    """
    stats = None

    train_df, test_df = get_train_n_test_data(tuner.project_name)
    train_ds, test_ds = df2ds(train_df), df2ds(test_df)

    for hp in tuner.get_best_hyperparameters(num_trials=num_models):
        new_entry = _create_model_stats(
            tuner,
            hp,
            stats=stats,
            max_epochs=max_epochs,
            num_runs=10,
            top_runs=5,
            batch_size=batch_size,
            patience=patience,
            verbose=verbose,
            train_ds=train_ds,
            test_ds=test_ds,
        )
        if stats is None:
            stats = new_entry
        else:
            stats = pd.concat([stats, new_entry]).reset_index(drop=True)

        try:
            display(stats.sort_values(f"{tuner.oracle.objective.name}_mean"))  # type: ignore
        # nosemgrep
        except Exception as e:  # nosec
            pass

    return stats.sort_values(f"{tuner.oracle.objective.name}_mean")  # type: ignore