Skip to content

iSOUPTreeRegressor

Incremental Structured Output Prediction Tree (iSOUP-Tree) for multi-target regression.

This is an implementation of the iSOUP-Tree proposed by A. Osojnik, P. Panov, and S. DΕΎeroski 1.

Parameters

  • grace_period (int) – defaults to 200

    Number of instances a leaf should observe between split attempts.

  • max_depth (int) – defaults to None

    The maximum depth a tree can reach. If None, the tree will grow indefinitely.

  • split_confidence (float) – defaults to 1e-07

    Allowed error in split decision, a value closer to 0 takes longer to decide.

  • tie_threshold (float) – defaults to 0.05

    Threshold below which a split will be forced to break ties.

  • leaf_prediction (str) – defaults to model

    Prediction mechanism used at leafs.
    - 'mean' - Target mean
    - 'model' - Uses the model defined in leaf_model
    - 'adaptive' - Chooses between 'mean' and 'model' dynamically

  • leaf_model (Union[base.Regressor, Dict]) – defaults to None

    The regression model(s) used to provide responses if leaf_prediction='model'. It can be either a regressor (in which case it is going to be replicated to all the targets) or a dictionary whose keys are target identifiers, and the values are instances of river.base.Regressor. If not provided, instances of river.linear_model.LinearRegression with the default hyperparameters are used for all the targets. If a dictionary is passed and not all target models are specified, copies from the first model match in the dictionary will be used to the remaining targets.

  • model_selector_decay (float) – defaults to 0.95

    The exponential decaying factor applied to the learning models' squared errors, that are monitored if leaf_prediction='adaptive'. Must be between 0 and 1. The closer to 1, the more importance is going to be given to past observations. On the other hand, if its value approaches 0, the recent observed errors are going to have more influence on the final decision.

  • nominal_attributes (list) – defaults to None

    List of Nominal attributes identifiers. If empty, then assume that all numeric attributes should be treated as continuous.

  • splitter (river.tree.splitter.base.Splitter) – defaults to None

    The Splitter or Attribute Observer (AO) used to monitor the class statistics of numeric features and perform splits. Splitters are available in the tree.splitter module. Different splitters are available for classification and regression tasks. Classification and regression splitters can be distinguished by their property is_target_class. This is an advanced option. Special care must be taken when choosing different splitters. By default, tree.splitter.EBSTSplitter is used if splitter is None.

  • min_samples_split (int) – defaults to 5

    The minimum number of samples every branch resulting from a split candidate must have to be considered valid.

  • binary_split (bool) – defaults to False

    If True, only allow binary splits.

  • max_size (float) – defaults to 500.0

    The max size of the tree, in Megabytes (MB).

  • memory_estimate_period (int) – defaults to 1000000

    Interval (number of processed instances) between memory consumption checks.

  • stop_mem_management (bool) – defaults to False

    If True, stop growing as soon as memory limit is hit.

  • remove_poor_attrs (bool) – defaults to False

    If True, disable poor attributes to reduce memory usage.

  • merit_preprune (bool) – defaults to True

    If True, enable merit-based tree pre-pruning.

Attributes

  • height

  • leaf_prediction

    Return the prediction strategy used by the tree at its leaves.

  • max_size

    Max allowed size tree can reach (in MB).

  • n_active_leaves

  • n_branches

  • n_inactive_leaves

  • n_leaves

  • n_nodes

  • split_criterion

    Return a string with the name of the split criterion being used by the tree.

  • summary

    Collect metrics corresponding to the current status of the tree in a string buffer.

Examples

>>> import numbers
>>> from river import compose
>>> from river import datasets
>>> from river import evaluate
>>> from river import linear_model
>>> from river import metrics
>>> from river import preprocessing
>>> from river import tree

>>> dataset = datasets.SolarFlare()

>>> num = compose.SelectType(numbers.Number) | preprocessing.MinMaxScaler()
>>> cat = compose.SelectType(str) | preprocessing.OneHotEncoder(sparse=False)

>>> model = tree.iSOUPTreeRegressor(
...     grace_period=100,
...     leaf_prediction='model',
...     leaf_model={
...         'c-class-flares': linear_model.LinearRegression(l2=0.02),
...         'm-class-flares': linear_model.PARegressor(),
...         'x-class-flares': linear_model.LinearRegression(l2=0.1)
...     }
... )

>>> pipeline = (num + cat) | model
>>> metric = metrics.multioutput.MicroAverage(metrics.MAE())

>>> evaluate.progressive_val_score(dataset, pipeline, metric)
MicroAverage(MAE): 0.426177

Methods

clone

Return a fresh estimator with the same parameters.

The clone has the same parameters but has not been updated with any data. This works by looking at the parameters from the class signature. Each parameter is either - recursively cloned if it's a River classes. - deep-copied via copy.deepcopy if not. If the calling object is stochastic (i.e. it accepts a seed parameter) and has not been seeded, then the clone will not be idempotent. Indeed, this method's purpose if simply to return a new instance with the same input parameters.

debug_one

Print an explanation of how x is predicted.

Parameters

  • x (dict)

Returns

typing.Union[str, NoneType]: A representation of the path followed by the tree to predict x; None if

draw

Draw the tree using the graphviz library.

Since the tree is drawn without passing incoming samples, classification trees will show the majority class in their leaves, whereas regression trees will use the target mean.

Parameters

  • max_depth (int) – defaults to None
    The maximum depth a tree can reach. If None, the tree will grow indefinitely.
learn_one

Incrementally train the model with one sample.

Training tasks: * If the tree is empty, create a leaf node as the root. * If the tree is already initialized, find the corresponding leaf for the instance and update the leaf node statistics. * If growth is allowed and the number of instances that the leaf has observed between split attempts exceed the grace period then attempt to split.

Parameters

  • x (dict)
  • y (Dict[Hashable, numbers.Number])
  • sample_weight (float) – defaults to 1.0
predict_one

Predict the target values for a given instance.

Parameters

  • x (dict)

Returns

typing.Dict[typing.Hashable, numbers.Number]: dict

to_dataframe

Return a representation of the current tree structure organized in a pandas.DataFrame object.

In case the tree is empty or it only contains a single node (a leaf), None is returned.

Returns

df

References

  1. AljaΕΎ Osojnik, Panče Panov, and SaΕ‘o DΕΎeroski. "Tree-based methods for online multi-target regression." Journal of Intelligent Information Systems 50.2 (2018): 315-339. ↩