You want to do stacking if more than one algorithm shall be applied, which equals to an AND-Operation.

The PHOTONAI Stack delivers the data to all of the entailed PipelineElements and the transformations or predictions are afterwards horizontally concatenated.

In this way you can preprocess data in different ways and collect the resulting information to create a new feature matrix. Additionally, you can train several learning algorithms with the same data in an ensemble-like fashion and concatenate their predictions to a prediction matrix on which you can apply further processing like voting strategies.

from sklearn.datasets import load_breast_cancer
from sklearn.model_selection import KFold

from photonai.base import Hyperpipe, PipelineElement, Stack, OutputSettings
from photonai.optimization import FloatRange, IntegerRange

X, y = load_breast_cancer(return_X_y=True)

my_pipe = Hyperpipe('basic_stack_pipe',
                    optimizer_params={'n_configurations': 5},
                    metrics=['accuracy', 'precision', 'recall'],

my_pipe += PipelineElement('StandardScaler')

tree = PipelineElement('DecisionTreeClassifier',
                       hyperparameters={'criterion': ['gini'],
                                         'min_samples_split': IntegerRange(2, 4)})

svc = PipelineElement('LinearSVC',
                      hyperparameters={'C': FloatRange(0.5, 25)})

# for a stack that includes estimators you can choose whether predict or predict_proba is called for all estimators
# in case only some implement predict_proba, predict is called for the remaining estimators
my_pipe += Stack('final_stack', [tree, svc], use_probabilities=True)

my_pipe += PipelineElement('LinearSVC'), y)