from sklearn.tree.tree import DecisionTreeClassifier as SKLModel
import lale.helpers
import lale.operators
from numpy import nan, inf
[docs]class DecisionTreeClassifierImpl():
def __init__(self, criterion='gini', splitter='best', max_depth=None, min_samples_split=2, min_samples_leaf=1, min_weight_fraction_leaf=0.0, max_features=None, random_state=None, max_leaf_nodes=None, min_impurity_decrease=0.0, min_impurity_split=None, class_weight='balanced', presort=False):
self._hyperparams = {
'criterion': criterion,
'splitter': splitter,
'max_depth': max_depth,
'min_samples_split': min_samples_split,
'min_samples_leaf': min_samples_leaf,
'min_weight_fraction_leaf': min_weight_fraction_leaf,
'max_features': max_features,
'random_state': random_state,
'max_leaf_nodes': max_leaf_nodes,
'min_impurity_decrease': min_impurity_decrease,
'min_impurity_split': min_impurity_split,
'class_weight': class_weight,
'presort': presort}
[docs] def fit(self, X, y=None):
self._sklearn_model = SKLModel(**self._hyperparams)
if (y is not None):
self._sklearn_model.fit(X, y)
else:
self._sklearn_model.fit(X)
return self
[docs] def predict(self, X):
return self._sklearn_model.predict(X)
[docs] def predict_proba(self, X):
return self._sklearn_model.predict_proba(X)
_hyperparams_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'inherited docstring for DecisionTreeClassifier A decision tree classifier.',
'allOf': [{
'type': 'object',
'required': ['criterion', 'splitter', 'max_depth', 'min_samples_split', 'min_samples_leaf', 'min_weight_fraction_leaf', 'max_features', 'random_state', 'max_leaf_nodes', 'min_impurity_decrease', 'min_impurity_split', 'class_weight', 'presort'],
'relevantToOptimizer': ['criterion', 'splitter', 'max_depth', 'min_samples_split', 'min_samples_leaf', 'max_features'],
'additionalProperties': False,
'properties': {
'criterion': {
'enum': ['entropy', 'gini'],
'default': 'gini',
'description': 'The function to measure the quality of a split. Supported criteria are'},
'splitter': {
'enum': ['random', 'best'],
'default': 'best',
'description': 'The strategy used to choose the split at each node. Supported'},
'max_depth': {
'anyOf': [{
'type': 'integer',
'minimumForOptimizer': 3,
'maximumForOptimizer': 5,
'distribution': 'uniform'}, {
'enum': [None]}],
'default': None,
'description': 'The maximum depth of the tree. If None, then nodes are expanded until'},
'min_samples_split': {
'anyOf': [{
'type': 'integer',
'minimumForOptimizer': 2,
'maximumForOptimizer': 5,
'distribution': 'uniform'}, {
'type': 'number',
'forOptimizer': False}],
'default': 2,
'description': 'The minimum number of samples required to split an internal node:'},
'min_samples_leaf': {
'anyOf': [{
'type': 'integer',
'minimumForOptimizer': 1,
'maximumForOptimizer': 5,
'distribution': 'uniform'}, {
'type': 'number',
'forOptimizer': False}],
'default': 1,
'description': 'The minimum number of samples required to be at a leaf node.'},
'min_weight_fraction_leaf': {
'type': 'number',
'default': 0.0,
'description': 'The minimum weighted fraction of the sum total of weights (of all'},
'max_features': {
'anyOf': [{
'type': 'integer',
'forOptimizer': False}, {
'type': 'number',
'minimumForOptimizer': 0.01,
'maximumForOptimizer': 1.0,
'distribution': 'uniform'}, {
'type': 'string',
'forOptimizer': False}, {
'enum': [None]}],
'default': None,
'description': 'The number of features to consider when looking for the best split:'},
'random_state': {
'anyOf': [{
'type': 'integer'}, {
'type': 'object'}, {
'enum': [None]}],
'default': None,
'description': 'If int, random_state is the seed used by the random number generator;'},
'max_leaf_nodes': {
'anyOf': [{
'type': 'integer'}, {
'enum': [None]}],
'default': None,
'description': 'Grow a tree with ``max_leaf_nodes`` in best-first fashion.'},
'min_impurity_decrease': {
'type': 'number',
'default': 0.0,
'description': 'A node will be split if this split induces a decrease of the impurity'},
'min_impurity_split': {
'anyOf': [{
'type': 'number'}, {
'enum': [None]}],
'default': None,
'description': 'Threshold for early stopping in tree growth. A node will split'},
'class_weight': {
'XXX TODO XXX': 'dict, list of dicts, "balanced" or None, default=None',
'description': 'Weights associated with classes in the form ``{class_label: weight}``.',
'enum': ['balanced'],
'default': 'balanced'},
'presort': {
'type': 'boolean',
'default': False,
'description': 'Whether to presort the data to speed up the finding of best splits in'},
}}, {
'XXX TODO XXX': 'Parameter: min_samples_leaf > only be considered if it leaves at least min_samples_leaf training samples in each of the left and right branches'}],
}
_input_fit_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'Build a decision tree classifier from the training set (X, y).',
'type': 'object',
'properties': {
'X': {
'anyOf': [{
'type': 'array',
'items': {
'XXX TODO XXX': 'item type'},
'XXX TODO XXX': 'array-like or sparse matrix, shape = [n_samples, n_features]'}, {
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}],
'description': 'The training input samples. Internally, it will be converted to'},
'y': {
'anyOf': [{
'type': 'array',
'items': {
'type': 'number'},
}, {
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}],
'description': 'The target values (class labels) as integers or strings.'},
'sample_weight': {
'anyOf': [{
'type': 'array',
'items': {
'type': 'number'},
}, {
'enum': [None]}],
'description': 'Sample weights. If None, then samples are equally weighted. Splits'},
'check_input': {
'type': 'boolean',
'default': True,
'description': 'Allow to bypass several input checking.'},
'X_idx_sorted': {
'anyOf': [{
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}, {
'enum': [None]}],
'default': None,
'description': 'The indexes of the sorted training input samples. If many tree'},
},
}
_input_predict_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'Predict class or regression value for X.',
'type': 'object',
'properties': {
'X': {
'anyOf': [{
'type': 'array',
'items': {
'XXX TODO XXX': 'item type'},
'XXX TODO XXX': 'array-like or sparse matrix of shape = [n_samples, n_features]'}, {
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}],
'description': 'The input samples. Internally, it will be converted to'},
'check_input': {
'type': 'boolean',
'default': True,
'description': 'Allow to bypass several input checking.'},
},
}
_output_predict_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'The predicted classes, or the predict values.',
'anyOf': [{
'type': 'array',
'items': {
'type': 'number'},
}, {
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}],
}
_input_predict_proba_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'Predict class probabilities of the input samples X.',
'type': 'object',
'properties': {
'X': {
'anyOf': [{
'type': 'array',
'items': {
'XXX TODO XXX': 'item type'},
'XXX TODO XXX': 'array-like or sparse matrix of shape = [n_samples, n_features]'}, {
'type': 'array',
'items': {
'type': 'array',
'items': {
'type': 'number'},
}}],
'description': 'The input samples. Internally, it will be converted to'},
'check_input': {
'type': 'boolean',
'description': 'Run check_array on X.'},
},
}
_output_predict_proba_schema = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'such arrays if n_outputs > 1.',
'XXX TODO XXX': 'array of shape = [n_samples, n_classes], or a list of n_outputs',
}
_combined_schemas = {
'$schema': 'http://json-schema.org/draft-04/schema#',
'description': 'Combined schema for expected data and hyperparameters.',
'type': 'object',
'tags': {
'pre': [],
'op': ['estimator'],
'post': []},
'properties': {
'hyperparams': _hyperparams_schema,
'input_fit': _input_fit_schema,
'input_predict': _input_predict_schema,
'output_predict': _output_predict_schema,
'input_predict_proba': _input_predict_proba_schema,
'output_predict_proba': _output_predict_proba_schema},
}
if (__name__ == '__main__'):
lale.helpers.validate_is_schema(_combined_schemas)
DecisionTreeClassifier = lale.operators.make_operator(DecisionTreeClassifierImpl, _combined_schemas)