iSOUPTreeRegressor¶
Incremental Structured Output Prediction Tree (iSOUPTree) for multitarget regression.
This is an implementation of the iSOUPTree proposed by A. Osojnik, P. Panov, and S. Dลพeroski ^{1}.
Parameters¶

grace_period
Type โ int
Default โ
200
Number of instances a leaf should observe between split attempts.

max_depth
Type โ int  None
Default โ
None
The maximum depth a tree can reach. If
None
, the tree will grow indefinitely. 
delta
Type โ float
Default โ
1e07
Allowed error in split decision, a value closer to 0 takes longer to decide.

tau
Type โ float
Default โ
0.05
Threshold below which a split will be forced to break ties.

leaf_prediction
Type โ str
Default โ
adaptive
Prediction mechanism used at leafs.
 'mean'  Target mean
 'model'  Uses the model defined inleaf_model
 'adaptive'  Chooses between 'mean' and 'model' dynamically 
leaf_model
Type โ base.Regressor  dict  None
Default โ
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 ofbase.Regressor
.If not provided, instances of [
linear_model.LinearRegression`](../../linearmodel/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
Type โ float
Default โ
0.95
The exponential decaying factor applied to the learning models' squared errors, that are monitored if
leaf_prediction='adaptive'
. Must be between0
and1
. The closer to1
, the more importance is going to be given to past observations. On the other hand, if its value approaches0
, the recent observed errors are going to have more influence on the final decision. 
nominal_attributes
Type โ list  None
Default โ
None
List of Nominal attributes identifiers. If empty, then assume that all numeric attributes should be treated as continuous.

splitter
Type โ Splitter  None
Default โ
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 propertyis_target_class
. This is an advanced option. Special care must be taken when choosing different splitters. By default,tree.splitter.TEBSTSplitter
is used ifsplitter
isNone
. 
min_samples_split
Type โ int
Default โ
5
The minimum number of samples every branch resulting from a split candidate must have to be considered valid.

binary_split
Type โ bool
Default โ
False
If True, only allow binary splits.

max_size
Type โ float
Default โ
500.0
The max size of the tree, in Megabytes (MB).

memory_estimate_period
Type โ int
Default โ
1000000
Interval (number of processed instances) between memory consumption checks.

stop_mem_management
Type โ bool
Default โ
False
If True, stop growing as soon as memory limit is hit.

remove_poor_attrs
Type โ bool
Default โ
False
If True, disable poor attributes to reduce memory usage.

merit_preprune
Type โ bool
Default โ
True
If True, enable meritbased tree prepruning.
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()
model = tree.iSOUPTreeRegressor(
grace_period=100,
leaf_prediction='model',
leaf_model={
'cclassflares': linear_model.LinearRegression(l2=0.02),
'mclassflares': linear_model.PARegressor(),
'xclassflares': 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¶
debug_one
Print an explanation of how x
is predicted.
Parameters
 x โ 'dict'
Returns
str  None: 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 โ 'int  None' โ defaults to
None
The maximum depth a tree can reach. IfNone
, 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
 y
 w โ 'float' โ defaults to
1.0
predict_one
Predict the target value using one of the leaf prediction strategies.
Parameters
 x
Returns
Predicted target value.
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

Aljaลพ Osojnik, Panฤe Panov, and Saลกo Dลพeroski. "Treebased methods for online multitarget regression." Journal of Intelligent Information Systems 50.2 (2018): 315339. ↩