Coverage for pyspark/ml/regression.py : 84%
Hot-keys on this page
r m x p toggle line displays
j k next/prev highlighted chunk
0 (zero) top of page
1 (one) first highlighted chunk
|
# # Licensed to the Apache Software Foundation (ASF) under one or more # contributor license agreements. See the NOTICE file distributed with # this work for additional information regarding copyright ownership. # The ASF licenses this file to You under the Apache License, Version 2.0 # (the "License"); you may not use this file except in compliance with # the License. You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. #
Param, Params, TypeConverters, HasMaxIter, HasTol, HasFitIntercept, HasAggregationDepth, \ HasMaxBlockSizeInMB, HasRegParam, HasSolver, HasStepSize, HasSeed, HasElasticNetParam, \ HasStandardization, HasLoss, HasVarianceCol _TreeEnsembleModel, _RandomForestParams, _GBTParams, _TreeRegressorParams GeneralJavaMLWritable JavaPredictor, JavaPredictionModel, JavaWrapper
'DecisionTreeRegressor', 'DecisionTreeRegressionModel', 'GBTRegressor', 'GBTRegressionModel', 'GeneralizedLinearRegression', 'GeneralizedLinearRegressionModel', 'GeneralizedLinearRegressionSummary', 'GeneralizedLinearRegressionTrainingSummary', 'IsotonicRegression', 'IsotonicRegressionModel', 'LinearRegression', 'LinearRegressionModel', 'LinearRegressionSummary', 'LinearRegressionTrainingSummary', 'RandomForestRegressor', 'RandomForestRegressionModel', 'FMRegressor', 'FMRegressionModel']
""" Regressor for regression tasks.
.. versionadded:: 3.0.0 """
""" Model produced by a ``Regressor``.
.. versionadded:: 3.0.0 """
""" Java Regressor for regression tasks.
.. versionadded:: 3.0.0 """
""" Java Model produced by a ``_JavaRegressor``. To be mixed in with :class:`pyspark.ml.JavaModel`
.. versionadded:: 3.0.0 """
HasTol, HasFitIntercept, HasStandardization, HasWeightCol, HasSolver, HasAggregationDepth, HasLoss, HasMaxBlockSizeInMB): """ Params for :py:class:`LinearRegression` and :py:class:`LinearRegressionModel`.
.. versionadded:: 3.0.0 """
"options: auto, normal, l-bfgs.", typeConverter=TypeConverters.toString)
"options: squaredError, huber.", typeConverter=TypeConverters.toString)
"robustness. Must be > 1.0. Only valid when loss is huber", typeConverter=TypeConverters.toFloat)
maxBlockSizeInMB=0.0)
def getEpsilon(self): """ Gets the value of epsilon or its default value. """ return self.getOrDefault(self.epsilon)
""" Linear regression.
The learning objective is to minimize the specified loss function, with regularization. This supports two kinds of loss:
* squaredError (a.k.a squared loss) * huber (a hybrid of squared error for relatively small errors and absolute error for \ relatively large ones, and we estimate the scale parameter from training data)
This supports multiple types of regularization:
* none (a.k.a. ordinary least squares) * L2 (ridge regression) * L1 (Lasso) * L2 + L1 (elastic net)
.. versionadded:: 1.4.0
Notes ----- Fitting with huber loss only supports none and L2 regularization.
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, 2.0, Vectors.dense(1.0)), ... (0.0, 2.0, Vectors.sparse(1, [], []))], ["label", "weight", "features"]) >>> lr = LinearRegression(regParam=0.0, solver="normal", weightCol="weight") >>> lr.setMaxIter(5) LinearRegression... >>> lr.getMaxIter() 5 >>> lr.setRegParam(0.1) LinearRegression... >>> lr.getRegParam() 0.1 >>> lr.setRegParam(0.0) LinearRegression... >>> model = lr.fit(df) >>> model.setFeaturesCol("features") LinearRegressionModel... >>> model.setPredictionCol("newPrediction") LinearRegressionModel... >>> model.getMaxIter() 5 >>> model.getMaxBlockSizeInMB() 0.0 >>> test0 = spark.createDataFrame([(Vectors.dense(-1.0),)], ["features"]) >>> abs(model.predict(test0.head().features) - (-1.0)) < 0.001 True >>> abs(model.transform(test0).head().newPrediction - (-1.0)) < 0.001 True >>> abs(model.coefficients[0] - 1.0) < 0.001 True >>> abs(model.intercept - 0.0) < 0.001 True >>> test1 = spark.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"]) >>> abs(model.transform(test1).head().newPrediction - 1.0) < 0.001 True >>> lr.setParams(featuresCol="vector") LinearRegression... >>> lr_path = temp_path + "/lr" >>> lr.save(lr_path) >>> lr2 = LinearRegression.load(lr_path) >>> lr2.getMaxIter() 5 >>> model_path = temp_path + "/lr_model" >>> model.save(model_path) >>> model2 = LinearRegressionModel.load(model_path) >>> model.coefficients[0] == model2.coefficients[0] True >>> model.intercept == model2.intercept True >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True >>> model.numFeatures 1 >>> model.write().format("pmml").save(model_path + "_2") """
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, standardization=True, solver="auto", weightCol=None, aggregationDepth=2, loss="squaredError", epsilon=1.35, maxBlockSizeInMB=0.0): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, \ standardization=True, solver="auto", weightCol=None, aggregationDepth=2, \ loss="squaredError", epsilon=1.35, maxBlockSizeInMB=0.0) """ "org.apache.spark.ml.regression.LinearRegression", self.uid)
maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, standardization=True, solver="auto", weightCol=None, aggregationDepth=2, loss="squaredError", epsilon=1.35, maxBlockSizeInMB=0.0): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxIter=100, regParam=0.0, elasticNetParam=0.0, tol=1e-6, fitIntercept=True, \ standardization=True, solver="auto", weightCol=None, aggregationDepth=2, \ loss="squaredError", epsilon=1.35, maxBlockSizeInMB=0.0) Sets params for linear regression. """
def setEpsilon(self, value): """ Sets the value of :py:attr:`epsilon`. """ return self._set(epsilon=value)
""" Sets the value of :py:attr:`maxIter`. """
""" Sets the value of :py:attr:`regParam`. """
""" Sets the value of :py:attr:`tol`. """ return self._set(tol=value)
""" Sets the value of :py:attr:`elasticNetParam`. """ return self._set(elasticNetParam=value)
""" Sets the value of :py:attr:`fitIntercept`. """ return self._set(fitIntercept=value)
""" Sets the value of :py:attr:`standardization`. """ return self._set(standardization=value)
""" Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
""" Sets the value of :py:attr:`solver`. """ return self._set(solver=value)
""" Sets the value of :py:attr:`aggregationDepth`. """ return self._set(aggregationDepth=value)
""" Sets the value of :py:attr:`loss`. """ return self._set(lossType=value)
def setMaxBlockSizeInMB(self, value): """ Sets the value of :py:attr:`maxBlockSizeInMB`. """ return self._set(maxBlockSizeInMB=value)
JavaMLReadable, HasTrainingSummary): """ Model fitted by :class:`LinearRegression`.
.. versionadded:: 1.4.0 """
def coefficients(self): """ Model coefficients. """
def intercept(self): """ Model intercept. """
def scale(self): r""" The value by which :math:`\|y - X'w\|` is scaled down when loss is "huber", otherwise 1.0. """
def summary(self): """ Gets summary (residuals, MSE, r-squared ) of model on training set. An exception is thrown if `trainingSummary is None`. """ else: raise RuntimeError("No training summary available for this %s" % self.__class__.__name__)
""" Evaluates the model on a test dataset.
.. versionadded:: 2.0.0
Parameters ---------- dataset : :py:class:`pyspark.sql.DataFrame` Test dataset to evaluate model on, where dataset is an instance of :py:class:`pyspark.sql.DataFrame` """
""" Linear regression results evaluated on a dataset.
.. versionadded:: 2.0.0 """
def predictions(self): """ Dataframe outputted by the model's `transform` method. """
def predictionCol(self): """ Field in "predictions" which gives the predicted value of the label at each instance. """
def labelCol(self): """ Field in "predictions" which gives the true label of each instance. """
def featuresCol(self): """ Field in "predictions" which gives the features of each instance as a vector. """
def explainedVariance(self): r""" Returns the explained variance regression score. explainedVariance = :math:`1 - \frac{variance(y - \hat{y})}{variance(y)}`
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions.
For additional information see `Explained variation on Wikipedia \ <http://en.wikipedia.org/wiki/Explained_variation>`_ """
def meanAbsoluteError(self): """ Returns the mean absolute error, which is a risk function corresponding to the expected value of the absolute error loss or l1-norm loss.
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions. """
def meanSquaredError(self): """ Returns the mean squared error, which is a risk function corresponding to the expected value of the squared error loss or quadratic loss.
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions. """
def rootMeanSquaredError(self): """ Returns the root mean squared error, which is defined as the square root of the mean squared error.
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions. """
def r2(self): """ Returns R^2, the coefficient of determination.
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions.
See also `Wikipedia coefficient of determination \ <http://en.wikipedia.org/wiki/Coefficient_of_determination>`_ """
def r2adj(self): """ Returns Adjusted R^2, the adjusted coefficient of determination.
Notes ----- This ignores instance weights (setting all to 1.0) from `LinearRegression.weightCol`. This will change in later Spark versions.
`Wikipedia coefficient of determination, Adjusted R^2 \ <https://en.wikipedia.org/wiki/Coefficient_of_determination#Adjusted_R2>`_ """
def residuals(self): """ Residuals (label - predicted value) """
def numInstances(self): """ Number of instances in DataFrame predictions """
def degreesOfFreedom(self): """ Degrees of freedom. """
def devianceResiduals(self): """ The weighted residuals, the usual residuals rescaled by the square root of the instance weights. """
def coefficientStandardErrors(self): """ Standard error of estimated coefficients and intercept. This value is only available when using the "normal" solver.
If :py:attr:`LinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept.
.. versionadded:: 2.0.0
See Also -------- LinearRegression.solver """
def tValues(self): """ T-statistic of estimated coefficients and intercept. This value is only available when using the "normal" solver.
If :py:attr:`LinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept.
.. versionadded:: 2.0.0
See Also -------- LinearRegression.solver """
def pValues(self): """ Two-sided p-value of estimated coefficients and intercept. This value is only available when using the "normal" solver.
If :py:attr:`LinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept.
.. versionadded:: 2.0.0
See Also -------- LinearRegression.solver """
""" Linear regression training results. Currently, the training summary ignores the training weights except for the objective trace.
.. versionadded:: 2.0.0 """
def objectiveHistory(self): """ Objective function (scaled loss + regularization) at each iteration. This value is only available when using the "l-bfgs" solver.
.. versionadded:: 2.0.0
See Also -------- LinearRegression.solver """
def totalIterations(self): """ Number of training iterations until termination. This value is only available when using the "l-bfgs" solver.
.. versionadded:: 2.0.0
See Also -------- LinearRegression.solver """
""" Params for :py:class:`IsotonicRegression` and :py:class:`IsotonicRegressionModel`.
.. versionadded:: 3.0.0 """
Params._dummy(), "isotonic", "whether the output sequence should be isotonic/increasing (true) or" + "antitonic/decreasing (false).", typeConverter=TypeConverters.toBoolean) Params._dummy(), "featureIndex", "The index of the feature if featuresCol is a vector column, no effect otherwise.", typeConverter=TypeConverters.toInt)
""" Gets the value of isotonic or its default value. """
""" Gets the value of featureIndex or its default value. """
JavaMLWritable, JavaMLReadable): """ Currently implemented using parallelized pool adjacent violators algorithm. Only univariate (single feature) algorithm supported.
.. versionadded:: 1.6.0
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(1.0)), ... (0.0, Vectors.sparse(1, [], []))], ["label", "features"]) >>> ir = IsotonicRegression() >>> model = ir.fit(df) >>> model.setFeaturesCol("features") IsotonicRegressionModel... >>> model.numFeatures 1 >>> test0 = spark.createDataFrame([(Vectors.dense(-1.0),)], ["features"]) >>> model.transform(test0).head().prediction 0.0 >>> model.predict(test0.head().features[model.getFeatureIndex()]) 0.0 >>> model.boundaries DenseVector([0.0, 1.0]) >>> ir_path = temp_path + "/ir" >>> ir.save(ir_path) >>> ir2 = IsotonicRegression.load(ir_path) >>> ir2.getIsotonic() True >>> model_path = temp_path + "/ir_model" >>> model.save(model_path) >>> model2 = IsotonicRegressionModel.load(model_path) >>> model.boundaries == model2.boundaries True >>> model.predictions == model2.predictions True >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True """ weightCol=None, isotonic=True, featureIndex=0): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ weightCol=None, isotonic=True, featureIndex=0): """ "org.apache.spark.ml.regression.IsotonicRegression", self.uid)
weightCol=None, isotonic=True, featureIndex=0): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ weightCol=None, isotonic=True, featureIndex=0): Set the params for IsotonicRegression. """
""" Sets the value of :py:attr:`isotonic`. """ return self._set(isotonic=value)
""" Sets the value of :py:attr:`featureIndex`. """ return self._set(featureIndex=value)
def setFeaturesCol(self, value): """ Sets the value of :py:attr:`featuresCol`. """ return self._set(featuresCol=value)
def setPredictionCol(self, value): """ Sets the value of :py:attr:`predictionCol`. """ return self._set(predictionCol=value)
def setLabelCol(self, value): """ Sets the value of :py:attr:`labelCol`. """ return self._set(labelCol=value)
def setWeightCol(self, value): """ Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
JavaMLReadable): """ Model fitted by :class:`IsotonicRegression`.
.. versionadded:: 1.6.0 """
def setFeaturesCol(self, value): """ Sets the value of :py:attr:`featuresCol`. """
def setPredictionCol(self, value): """ Sets the value of :py:attr:`predictionCol`. """ return self._set(predictionCol=value)
""" Sets the value of :py:attr:`featureIndex`. """ return self._set(featureIndex=value)
def boundaries(self): """ Boundaries in increasing order for which predictions are known. """
def predictions(self): """ Predictions associated with the boundaries at the same index, monotone because of isotonic regression. """
def numFeatures(self): """ Returns the number of features the model was trained on. If unknown, returns -1 """
def predict(self, value): """ Predict label for the given features. """
""" Params for :py:class:`DecisionTreeRegressor` and :py:class:`DecisionTreeRegressionModel`.
.. versionadded:: 3.0.0 """
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", leafCol="", minWeightFractionPerNode=0.0)
JavaMLReadable): """ `Decision tree <http://en.wikipedia.org/wiki/Decision_tree_learning>`_ learning algorithm for regression. It supports both continuous and categorical features.
.. versionadded:: 1.4.0
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(1.0)), ... (0.0, Vectors.sparse(1, [], []))], ["label", "features"]) >>> dt = DecisionTreeRegressor(maxDepth=2) >>> dt.setVarianceCol("variance") DecisionTreeRegressor... >>> model = dt.fit(df) >>> model.getVarianceCol() 'variance' >>> model.setLeafCol("leafId") DecisionTreeRegressionModel... >>> model.depth 1 >>> model.numNodes 3 >>> model.featureImportances SparseVector(1, {0: 1.0}) >>> model.numFeatures 1 >>> test0 = spark.createDataFrame([(Vectors.dense(-1.0),)], ["features"]) >>> model.predict(test0.head().features) 0.0 >>> result = model.transform(test0).head() >>> result.prediction 0.0 >>> model.predictLeaf(test0.head().features) 0.0 >>> result.leafId 0.0 >>> test1 = spark.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"]) >>> model.transform(test1).head().prediction 1.0 >>> dtr_path = temp_path + "/dtr" >>> dt.save(dtr_path) >>> dt2 = DecisionTreeRegressor.load(dtr_path) >>> dt2.getMaxDepth() 2 >>> model_path = temp_path + "/dtr_model" >>> model.save(model_path) >>> model2 = DecisionTreeRegressionModel.load(model_path) >>> model.numNodes == model2.numNodes True >>> model.depth == model2.depth True >>> model.transform(test1).head().variance 0.0 >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True >>> df3 = spark.createDataFrame([ ... (1.0, 0.2, Vectors.dense(1.0)), ... (1.0, 0.8, Vectors.dense(1.0)), ... (0.0, 1.0, Vectors.sparse(1, [], []))], ["label", "weight", "features"]) >>> dt3 = DecisionTreeRegressor(maxDepth=2, weightCol="weight", varianceCol="variance") >>> model3 = dt3.fit(df3) >>> print(model3.toDebugString) DecisionTreeRegressionModel...depth=1, numNodes=3... """
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", seed=None, varianceCol=None, weightCol=None, leafCol="", minWeightFractionPerNode=0.0): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \ impurity="variance", seed=None, varianceCol=None, weightCol=None, \ leafCol="", minWeightFractionPerNode=0.0) """ "org.apache.spark.ml.regression.DecisionTreeRegressor", self.uid)
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", seed=None, varianceCol=None, weightCol=None, leafCol="", minWeightFractionPerNode=0.0): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \ impurity="variance", seed=None, varianceCol=None, weightCol=None, \ leafCol="", minWeightFractionPerNode=0.0) Sets params for the DecisionTreeRegressor. """
def setMaxDepth(self, value): """ Sets the value of :py:attr:`maxDepth`. """ return self._set(maxDepth=value)
def setMaxBins(self, value): """ Sets the value of :py:attr:`maxBins`. """ return self._set(maxBins=value)
def setMinInstancesPerNode(self, value): """ Sets the value of :py:attr:`minInstancesPerNode`. """ return self._set(minInstancesPerNode=value)
def setMinWeightFractionPerNode(self, value): """ Sets the value of :py:attr:`minWeightFractionPerNode`. """ return self._set(minWeightFractionPerNode=value)
def setMinInfoGain(self, value): """ Sets the value of :py:attr:`minInfoGain`. """ return self._set(minInfoGain=value)
def setMaxMemoryInMB(self, value): """ Sets the value of :py:attr:`maxMemoryInMB`. """ return self._set(maxMemoryInMB=value)
def setCacheNodeIds(self, value): """ Sets the value of :py:attr:`cacheNodeIds`. """ return self._set(cacheNodeIds=value)
def setImpurity(self, value): """ Sets the value of :py:attr:`impurity`. """ return self._set(impurity=value)
def setCheckpointInterval(self, value): """ Sets the value of :py:attr:`checkpointInterval`. """ return self._set(checkpointInterval=value)
""" Sets the value of :py:attr:`seed`. """ return self._set(seed=value)
def setWeightCol(self, value): """ Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
def setVarianceCol(self, value): """ Sets the value of :py:attr:`varianceCol`. """
_JavaRegressionModel, _DecisionTreeModel, _DecisionTreeRegressorParams, JavaMLWritable, JavaMLReadable ): """ Model fitted by :class:`DecisionTreeRegressor`.
.. versionadded:: 1.4.0 """
def setVarianceCol(self, value): """ Sets the value of :py:attr:`varianceCol`. """ return self._set(varianceCol=value)
def featureImportances(self): """ Estimate of the importance of each feature.
This generalizes the idea of "Gini" importance to other losses, following the explanation of Gini importance from "Random Forests" documentation by Leo Breiman and Adele Cutler, and following the implementation from scikit-learn.
This feature importance is calculated as follows: - importance(feature j) = sum (over nodes which split on feature j) of the gain, where gain is scaled by the number of instances passing through node - Normalize importances for tree to sum to 1.
.. versionadded:: 2.0.0
Notes ----- Feature importance for single decision trees can have high variance due to correlated predictor variables. Consider using a :py:class:`RandomForestRegressor` to determine feature importance instead. """
""" Params for :py:class:`RandomForestRegressor` and :py:class:`RandomForestRegressionModel`.
.. versionadded:: 3.0.0 """
maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", subsamplingRate=1.0, numTrees=20, featureSubsetStrategy="auto", leafCol="", minWeightFractionPerNode=0.0, bootstrap=True)
JavaMLReadable): """ `Random Forest <http://en.wikipedia.org/wiki/Random_forest>`_ learning algorithm for regression. It supports both continuous and categorical features.
.. versionadded:: 1.4.0
Examples -------- >>> from numpy import allclose >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(1.0)), ... (0.0, Vectors.sparse(1, [], []))], ["label", "features"]) >>> rf = RandomForestRegressor(numTrees=2, maxDepth=2) >>> rf.getMinWeightFractionPerNode() 0.0 >>> rf.setSeed(42) RandomForestRegressor... >>> model = rf.fit(df) >>> model.getBootstrap() True >>> model.getSeed() 42 >>> model.setLeafCol("leafId") RandomForestRegressionModel... >>> model.featureImportances SparseVector(1, {0: 1.0}) >>> allclose(model.treeWeights, [1.0, 1.0]) True >>> test0 = spark.createDataFrame([(Vectors.dense(-1.0),)], ["features"]) >>> model.predict(test0.head().features) 0.0 >>> model.predictLeaf(test0.head().features) DenseVector([0.0, 0.0]) >>> result = model.transform(test0).head() >>> result.prediction 0.0 >>> result.leafId DenseVector([0.0, 0.0]) >>> model.numFeatures 1 >>> model.trees [DecisionTreeRegressionModel...depth=..., DecisionTreeRegressionModel...] >>> model.getNumTrees 2 >>> test1 = spark.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"]) >>> model.transform(test1).head().prediction 0.5 >>> rfr_path = temp_path + "/rfr" >>> rf.save(rfr_path) >>> rf2 = RandomForestRegressor.load(rfr_path) >>> rf2.getNumTrees() 2 >>> model_path = temp_path + "/rfr_model" >>> model.save(model_path) >>> model2 = RandomForestRegressionModel.load(model_path) >>> model.featureImportances == model2.featureImportances True >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True """
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, featureSubsetStrategy="auto", leafCol="", minWeightFractionPerNode=0.0, weightCol=None, bootstrap=True): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \ impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, \ featureSubsetStrategy="auto", leafCol=", minWeightFractionPerNode=0.0", \ weightCol=None, bootstrap=True) """ "org.apache.spark.ml.regression.RandomForestRegressor", self.uid)
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, featureSubsetStrategy="auto", leafCol="", minWeightFractionPerNode=0.0, weightCol=None, bootstrap=True): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, checkpointInterval=10, \ impurity="variance", subsamplingRate=1.0, seed=None, numTrees=20, \ featureSubsetStrategy="auto", leafCol="", minWeightFractionPerNode=0.0, \ weightCol=None, bootstrap=True) Sets params for linear regression. """
""" Sets the value of :py:attr:`maxDepth`. """ return self._set(maxDepth=value)
""" Sets the value of :py:attr:`maxBins`. """ return self._set(maxBins=value)
""" Sets the value of :py:attr:`minInstancesPerNode`. """ return self._set(minInstancesPerNode=value)
""" Sets the value of :py:attr:`minInfoGain`. """ return self._set(minInfoGain=value)
""" Sets the value of :py:attr:`maxMemoryInMB`. """ return self._set(maxMemoryInMB=value)
""" Sets the value of :py:attr:`cacheNodeIds`. """ return self._set(cacheNodeIds=value)
def setImpurity(self, value): """ Sets the value of :py:attr:`impurity`. """ return self._set(impurity=value)
def setNumTrees(self, value): """ Sets the value of :py:attr:`numTrees`. """ return self._set(numTrees=value)
def setBootstrap(self, value): """ Sets the value of :py:attr:`bootstrap`. """ return self._set(bootstrap=value)
def setSubsamplingRate(self, value): """ Sets the value of :py:attr:`subsamplingRate`. """ return self._set(subsamplingRate=value)
def setFeatureSubsetStrategy(self, value): """ Sets the value of :py:attr:`featureSubsetStrategy`. """ return self._set(featureSubsetStrategy=value)
""" Sets the value of :py:attr:`checkpointInterval`. """ return self._set(checkpointInterval=value)
""" Sets the value of :py:attr:`seed`. """
def setWeightCol(self, value): """ Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
def setMinWeightFractionPerNode(self, value): """ Sets the value of :py:attr:`minWeightFractionPerNode`. """ return self._set(minWeightFractionPerNode=value)
_JavaRegressionModel, _TreeEnsembleModel, _RandomForestRegressorParams, JavaMLWritable, JavaMLReadable ): """ Model fitted by :class:`RandomForestRegressor`.
.. versionadded:: 1.4.0 """
def trees(self): """Trees in this ensemble. Warning: These have null parent Estimators."""
def featureImportances(self): """ Estimate of the importance of each feature.
Each feature's importance is the average of its importance across all trees in the ensemble The importance vector is normalized to sum to 1. This method is suggested by Hastie et al. (Hastie, Tibshirani, Friedman. "The Elements of Statistical Learning, 2nd Edition." 2001.) and follows the implementation from scikit-learn.
.. versionadded:: 2.0.0
Examples -------- DecisionTreeRegressionModel.featureImportances """
""" Params for :py:class:`GBTRegressor` and :py:class:`GBTRegressorModel`.
.. versionadded:: 3.0.0 """
"Loss function which GBT tries to minimize (case-insensitive). " + "Supported options: " + ", ".join(supportedLossTypes), typeConverter=TypeConverters.toString)
maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1, impurity="variance", featureSubsetStrategy="all", validationTol=0.01, leafCol="", minWeightFractionPerNode=0.0)
def getLossType(self): """ Gets the value of lossType or its default value. """ return self.getOrDefault(self.lossType)
""" `Gradient-Boosted Trees (GBTs) <http://en.wikipedia.org/wiki/Gradient_boosting>`_ learning algorithm for regression. It supports both continuous and categorical features.
.. versionadded:: 1.4.0
Examples -------- >>> from numpy import allclose >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(1.0)), ... (0.0, Vectors.sparse(1, [], []))], ["label", "features"]) >>> gbt = GBTRegressor(maxDepth=2, seed=42, leafCol="leafId") >>> gbt.setMaxIter(5) GBTRegressor... >>> gbt.setMinWeightFractionPerNode(0.049) GBTRegressor... >>> gbt.getMaxIter() 5 >>> print(gbt.getImpurity()) variance >>> print(gbt.getFeatureSubsetStrategy()) all >>> model = gbt.fit(df) >>> model.featureImportances SparseVector(1, {0: 1.0}) >>> model.numFeatures 1 >>> allclose(model.treeWeights, [1.0, 0.1, 0.1, 0.1, 0.1]) True >>> test0 = spark.createDataFrame([(Vectors.dense(-1.0),)], ["features"]) >>> model.predict(test0.head().features) 0.0 >>> model.predictLeaf(test0.head().features) DenseVector([0.0, 0.0, 0.0, 0.0, 0.0]) >>> result = model.transform(test0).head() >>> result.prediction 0.0 >>> result.leafId DenseVector([0.0, 0.0, 0.0, 0.0, 0.0]) >>> test1 = spark.createDataFrame([(Vectors.sparse(1, [0], [1.0]),)], ["features"]) >>> model.transform(test1).head().prediction 1.0 >>> gbtr_path = temp_path + "gbtr" >>> gbt.save(gbtr_path) >>> gbt2 = GBTRegressor.load(gbtr_path) >>> gbt2.getMaxDepth() 2 >>> model_path = temp_path + "gbtr_model" >>> model.save(model_path) >>> model2 = GBTRegressionModel.load(model_path) >>> model.featureImportances == model2.featureImportances True >>> model.treeWeights == model2.treeWeights True >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True >>> model.trees [DecisionTreeRegressionModel...depth=..., DecisionTreeRegressionModel...] >>> validation = spark.createDataFrame([(0.0, Vectors.dense(-1.0))], ... ["label", "features"]) >>> model.evaluateEachIteration(validation, "squared") [0.0, 0.0, 0.0, 0.0, 0.0] >>> gbt = gbt.setValidationIndicatorCol("validationIndicator") >>> gbt.getValidationIndicatorCol() 'validationIndicator' >>> gbt.getValidationTol() 0.01 """
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1, seed=None, impurity="variance", featureSubsetStrategy="all", validationTol=0.01, validationIndicatorCol=None, leafCol="", minWeightFractionPerNode=0.0, weightCol=None): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, \ checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1, seed=None, \ impurity="variance", featureSubsetStrategy="all", validationTol=0.01, \ validationIndicatorCol=None, leafCol="", minWeightFractionPerNode=0.0, weightCol=None) """
maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1, seed=None, impurity="variance", featureSubsetStrategy="all", validationTol=0.01, validationIndicatorCol=None, leafCol="", minWeightFractionPerNode=0.0, weightCol=None): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ maxDepth=5, maxBins=32, minInstancesPerNode=1, minInfoGain=0.0, \ maxMemoryInMB=256, cacheNodeIds=False, subsamplingRate=1.0, \ checkpointInterval=10, lossType="squared", maxIter=20, stepSize=0.1, seed=None, \ impurity="variance", featureSubsetStrategy="all", validationTol=0.01, \ validationIndicatorCol=None, leafCol="", minWeightFractionPerNode=0.0, \ weightCol=None) Sets params for Gradient Boosted Tree Regression. """
def setMaxDepth(self, value): """ Sets the value of :py:attr:`maxDepth`. """ return self._set(maxDepth=value)
def setMaxBins(self, value): """ Sets the value of :py:attr:`maxBins`. """ return self._set(maxBins=value)
def setMinInstancesPerNode(self, value): """ Sets the value of :py:attr:`minInstancesPerNode`. """ return self._set(minInstancesPerNode=value)
def setMinInfoGain(self, value): """ Sets the value of :py:attr:`minInfoGain`. """ return self._set(minInfoGain=value)
def setMaxMemoryInMB(self, value): """ Sets the value of :py:attr:`maxMemoryInMB`. """ return self._set(maxMemoryInMB=value)
def setCacheNodeIds(self, value): """ Sets the value of :py:attr:`cacheNodeIds`. """ return self._set(cacheNodeIds=value)
def setImpurity(self, value): """ Sets the value of :py:attr:`impurity`. """ return self._set(impurity=value)
def setLossType(self, value): """ Sets the value of :py:attr:`lossType`. """ return self._set(lossType=value)
def setSubsamplingRate(self, value): """ Sets the value of :py:attr:`subsamplingRate`. """ return self._set(subsamplingRate=value)
def setFeatureSubsetStrategy(self, value): """ Sets the value of :py:attr:`featureSubsetStrategy`. """ return self._set(featureSubsetStrategy=value)
def setValidationIndicatorCol(self, value): """ Sets the value of :py:attr:`validationIndicatorCol`. """
def setMaxIter(self, value): """ Sets the value of :py:attr:`maxIter`. """
def setCheckpointInterval(self, value): """ Sets the value of :py:attr:`checkpointInterval`. """ return self._set(checkpointInterval=value)
def setSeed(self, value): """ Sets the value of :py:attr:`seed`. """ return self._set(seed=value)
def setStepSize(self, value): """ Sets the value of :py:attr:`stepSize`. """ return self._set(stepSize=value)
def setWeightCol(self, value): """ Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
def setMinWeightFractionPerNode(self, value): """ Sets the value of :py:attr:`minWeightFractionPerNode`. """
_JavaRegressionModel, _TreeEnsembleModel, _GBTRegressorParams, JavaMLWritable, JavaMLReadable ): """ Model fitted by :class:`GBTRegressor`.
.. versionadded:: 1.4.0 """
def featureImportances(self): """ Estimate of the importance of each feature.
Each feature's importance is the average of its importance across all trees in the ensemble The importance vector is normalized to sum to 1. This method is suggested by Hastie et al. (Hastie, Tibshirani, Friedman. "The Elements of Statistical Learning, 2nd Edition." 2001.) and follows the implementation from scikit-learn.
.. versionadded:: 2.0.0
Examples -------- DecisionTreeRegressionModel.featureImportances """
def trees(self): """Trees in this ensemble. Warning: These have null parent Estimators."""
""" Method to compute error or loss for every iteration of gradient boosting.
.. versionadded:: 2.4.0
Parameters ---------- dataset : :py:class:`pyspark.sql.DataFrame` Test dataset to evaluate model on, where dataset is an instance of :py:class:`pyspark.sql.DataFrame` loss : str The loss function used to compute error. Supported options: squared, absolute """
HasAggregationDepth, HasMaxBlockSizeInMB): """ Params for :py:class:`AFTSurvivalRegression` and :py:class:`AFTSurvivalRegressionModel`.
.. versionadded:: 3.0.0 """
Params._dummy(), "censorCol", "censor column name. The value of this column could be 0 or 1. " + "If the value is 1, it means the event has occurred i.e. " + "uncensored; otherwise censored.", typeConverter=TypeConverters.toString) Params._dummy(), "quantileProbabilities", "quantile probabilities array. Values of the quantile probabilities array " + "should be in the range (0, 1) and the array should be non-empty.", typeConverter=TypeConverters.toListFloat) Params._dummy(), "quantilesCol", "quantiles column name. This column will output quantiles of " + "corresponding quantileProbabilities if it is set.", typeConverter=TypeConverters.toString)
quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99], maxIter=100, tol=1E-6, maxBlockSizeInMB=0.0)
def getCensorCol(self): """ Gets the value of censorCol or its default value. """ return self.getOrDefault(self.censorCol)
def getQuantileProbabilities(self): """ Gets the value of quantileProbabilities or its default value. """ return self.getOrDefault(self.quantileProbabilities)
def getQuantilesCol(self): """ Gets the value of quantilesCol or its default value. """ return self.getOrDefault(self.quantilesCol)
JavaMLWritable, JavaMLReadable): """ Accelerated Failure Time (AFT) Model Survival Regression
Fit a parametric AFT survival regression model based on the Weibull distribution of the survival time.
Notes ----- For more information see Wikipedia page on `AFT Model <https://en.wikipedia.org/wiki/Accelerated_failure_time_model>`_
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(1.0), 1.0), ... (1e-40, Vectors.sparse(1, [], []), 0.0)], ["label", "features", "censor"]) >>> aftsr = AFTSurvivalRegression() >>> aftsr.setMaxIter(10) AFTSurvivalRegression... >>> aftsr.getMaxIter() 10 >>> aftsr.clear(aftsr.maxIter) >>> model = aftsr.fit(df) >>> model.getMaxBlockSizeInMB() 0.0 >>> model.setFeaturesCol("features") AFTSurvivalRegressionModel... >>> model.predict(Vectors.dense(6.3)) 1.0 >>> model.predictQuantiles(Vectors.dense(6.3)) DenseVector([0.0101, 0.0513, 0.1054, 0.2877, 0.6931, 1.3863, 2.3026, 2.9957, 4.6052]) >>> model.transform(df).show() +-------+---------+------+----------+ | label| features|censor|prediction| +-------+---------+------+----------+ | 1.0| [1.0]| 1.0| 1.0| |1.0E-40|(1,[],[])| 0.0| 1.0| +-------+---------+------+----------+ ... >>> aftsr_path = temp_path + "/aftsr" >>> aftsr.save(aftsr_path) >>> aftsr2 = AFTSurvivalRegression.load(aftsr_path) >>> aftsr2.getMaxIter() 100 >>> model_path = temp_path + "/aftsr_model" >>> model.save(model_path) >>> model2 = AFTSurvivalRegressionModel.load(model_path) >>> model.coefficients == model2.coefficients True >>> model.intercept == model2.intercept True >>> model.scale == model2.scale True >>> model.transform(df).take(1) == model2.transform(df).take(1) True
.. versionadded:: 1.6.0 """
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", quantileProbabilities=list([0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99]), # noqa: B005 quantilesCol=None, aggregationDepth=2, maxBlockSizeInMB=0.0): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", \ quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99], \ quantilesCol=None, aggregationDepth=2, maxBlockSizeInMB=0.0) """ "org.apache.spark.ml.regression.AFTSurvivalRegression", self.uid)
fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", quantileProbabilities=list([0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99]), # noqa: B005 quantilesCol=None, aggregationDepth=2, maxBlockSizeInMB=0.0): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ fitIntercept=True, maxIter=100, tol=1E-6, censorCol="censor", \ quantileProbabilities=[0.01, 0.05, 0.1, 0.25, 0.5, 0.75, 0.9, 0.95, 0.99], \ quantilesCol=None, aggregationDepth=2, maxBlockSizeInMB=0.0): """
def setCensorCol(self, value): """ Sets the value of :py:attr:`censorCol`. """ return self._set(censorCol=value)
def setQuantileProbabilities(self, value): """ Sets the value of :py:attr:`quantileProbabilities`. """ return self._set(quantileProbabilities=value)
def setQuantilesCol(self, value): """ Sets the value of :py:attr:`quantilesCol`. """ return self._set(quantilesCol=value)
def setMaxIter(self, value): """ Sets the value of :py:attr:`maxIter`. """
def setTol(self, value): """ Sets the value of :py:attr:`tol`. """ return self._set(tol=value)
def setFitIntercept(self, value): """ Sets the value of :py:attr:`fitIntercept`. """ return self._set(fitIntercept=value)
def setAggregationDepth(self, value): """ Sets the value of :py:attr:`aggregationDepth`. """ return self._set(aggregationDepth=value)
def setMaxBlockSizeInMB(self, value): """ Sets the value of :py:attr:`maxBlockSizeInMB`. """ return self._set(maxBlockSizeInMB=value)
JavaMLWritable, JavaMLReadable): """ Model fitted by :class:`AFTSurvivalRegression`.
.. versionadded:: 1.6.0 """
def setQuantileProbabilities(self, value): """ Sets the value of :py:attr:`quantileProbabilities`. """ return self._set(quantileProbabilities=value)
def setQuantilesCol(self, value): """ Sets the value of :py:attr:`quantilesCol`. """ return self._set(quantilesCol=value)
def coefficients(self): """ Model coefficients. """
def intercept(self): """ Model intercept. """
def scale(self): """ Model scale parameter. """
def predictQuantiles(self, features): """ Predicted Quantiles """
HasTol, HasRegParam, HasWeightCol, HasSolver, HasAggregationDepth): """ Params for :py:class:`GeneralizedLinearRegression` and :py:class:`GeneralizedLinearRegressionModel`.
.. versionadded:: 3.0.0 """
"the error distribution to be used in the model. Supported options: " + "gaussian (default), binomial, poisson, gamma and tweedie.", typeConverter=TypeConverters.toString) "relationship between the linear predictor and the mean of the distribution " + "function. Supported options: identity, log, inverse, logit, probit, cloglog " + "and sqrt.", typeConverter=TypeConverters.toString) "predictor) column name", typeConverter=TypeConverters.toString) "of the Tweedie distribution which characterizes the relationship " + "between the variance and mean of the distribution. Only applicable " + "for the Tweedie family. Supported values: 0 and [1, Inf).", typeConverter=TypeConverters.toFloat) "Only applicable to the Tweedie family.", typeConverter=TypeConverters.toFloat) "options: irls.", typeConverter=TypeConverters.toString) "or empty, we treat all instance offsets as 0.0", typeConverter=TypeConverters.toString)
variancePower=0.0, aggregationDepth=2)
def getFamily(self): """ Gets the value of family or its default value. """
def getLinkPredictionCol(self): """ Gets the value of linkPredictionCol or its default value. """ return self.getOrDefault(self.linkPredictionCol)
def getLink(self): """ Gets the value of link or its default value. """ return self.getOrDefault(self.link)
def getVariancePower(self): """ Gets the value of variancePower or its default value. """ return self.getOrDefault(self.variancePower)
def getLinkPower(self): """ Gets the value of linkPower or its default value. """ return self.getOrDefault(self.linkPower)
def getOffsetCol(self): """ Gets the value of offsetCol or its default value. """ return self.getOrDefault(self.offsetCol)
JavaMLWritable, JavaMLReadable): """ Generalized Linear Regression.
Fit a Generalized Linear Model specified by giving a symbolic description of the linear predictor (link function) and a description of the error distribution (family). It supports "gaussian", "binomial", "poisson", "gamma" and "tweedie" as family. Valid link functions for each family is listed below. The first link function of each family is the default one.
* "gaussian" -> "identity", "log", "inverse"
* "binomial" -> "logit", "probit", "cloglog"
* "poisson" -> "log", "identity", "sqrt"
* "gamma" -> "inverse", "identity", "log"
* "tweedie" -> power link function specified through "linkPower". \ The default link power in the tweedie family is 1 - variancePower.
.. versionadded:: 2.0.0
Notes ----- For more information see Wikipedia page on `GLM <https://en.wikipedia.org/wiki/Generalized_linear_model>`_
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> df = spark.createDataFrame([ ... (1.0, Vectors.dense(0.0, 0.0)), ... (1.0, Vectors.dense(1.0, 2.0)), ... (2.0, Vectors.dense(0.0, 0.0)), ... (2.0, Vectors.dense(1.0, 1.0)),], ["label", "features"]) >>> glr = GeneralizedLinearRegression(family="gaussian", link="identity", linkPredictionCol="p") >>> glr.setRegParam(0.1) GeneralizedLinearRegression... >>> glr.getRegParam() 0.1 >>> glr.clear(glr.regParam) >>> glr.setMaxIter(10) GeneralizedLinearRegression... >>> glr.getMaxIter() 10 >>> glr.clear(glr.maxIter) >>> model = glr.fit(df) >>> model.setFeaturesCol("features") GeneralizedLinearRegressionModel... >>> model.getMaxIter() 25 >>> model.getAggregationDepth() 2 >>> transformed = model.transform(df) >>> abs(transformed.head().prediction - 1.5) < 0.001 True >>> abs(transformed.head().p - 1.5) < 0.001 True >>> model.coefficients DenseVector([1.5..., -1.0...]) >>> model.numFeatures 2 >>> abs(model.intercept - 1.5) < 0.001 True >>> glr_path = temp_path + "/glr" >>> glr.save(glr_path) >>> glr2 = GeneralizedLinearRegression.load(glr_path) >>> glr.getFamily() == glr2.getFamily() True >>> model_path = temp_path + "/glr_model" >>> model.save(model_path) >>> model2 = GeneralizedLinearRegressionModel.load(model_path) >>> model.intercept == model2.intercept True >>> model.coefficients[0] == model2.coefficients[0] True >>> model.transform(df).take(1) == model2.transform(df).take(1) True """
family="gaussian", link=None, fitIntercept=True, maxIter=25, tol=1e-6, regParam=0.0, weightCol=None, solver="irls", linkPredictionCol=None, variancePower=0.0, linkPower=None, offsetCol=None, aggregationDepth=2): """ __init__(self, \\*, labelCol="label", featuresCol="features", predictionCol="prediction", \ family="gaussian", link=None, fitIntercept=True, maxIter=25, tol=1e-6, \ regParam=0.0, weightCol=None, solver="irls", linkPredictionCol=None, \ variancePower=0.0, linkPower=None, offsetCol=None, aggregationDepth=2) """ "org.apache.spark.ml.regression.GeneralizedLinearRegression", self.uid)
family="gaussian", link=None, fitIntercept=True, maxIter=25, tol=1e-6, regParam=0.0, weightCol=None, solver="irls", linkPredictionCol=None, variancePower=0.0, linkPower=None, offsetCol=None, aggregationDepth=2): """ setParams(self, \\*, labelCol="label", featuresCol="features", predictionCol="prediction", \ family="gaussian", link=None, fitIntercept=True, maxIter=25, tol=1e-6, \ regParam=0.0, weightCol=None, solver="irls", linkPredictionCol=None, \ variancePower=0.0, linkPower=None, offsetCol=None, aggregationDepth=2) Sets params for generalized linear regression. """
def setFamily(self, value): """ Sets the value of :py:attr:`family`. """ return self._set(family=value)
def setLinkPredictionCol(self, value): """ Sets the value of :py:attr:`linkPredictionCol`. """ return self._set(linkPredictionCol=value)
def setLink(self, value): """ Sets the value of :py:attr:`link`. """ return self._set(link=value)
def setVariancePower(self, value): """ Sets the value of :py:attr:`variancePower`. """ return self._set(variancePower=value)
def setLinkPower(self, value): """ Sets the value of :py:attr:`linkPower`. """
def setOffsetCol(self, value): """ Sets the value of :py:attr:`offsetCol`. """ return self._set(offsetCol=value)
def setMaxIter(self, value): """ Sets the value of :py:attr:`maxIter`. """
def setRegParam(self, value): """ Sets the value of :py:attr:`regParam`. """
def setTol(self, value): """ Sets the value of :py:attr:`tol`. """ return self._set(tol=value)
def setFitIntercept(self, value): """ Sets the value of :py:attr:`fitIntercept`. """ return self._set(fitIntercept=value)
def setWeightCol(self, value): """ Sets the value of :py:attr:`weightCol`. """ return self._set(weightCol=value)
def setSolver(self, value): """ Sets the value of :py:attr:`solver`. """ return self._set(solver=value)
def setAggregationDepth(self, value): """ Sets the value of :py:attr:`aggregationDepth`. """ return self._set(aggregationDepth=value)
JavaMLWritable, JavaMLReadable, HasTrainingSummary): """ Model fitted by :class:`GeneralizedLinearRegression`.
.. versionadded:: 2.0.0 """
def setLinkPredictionCol(self, value): """ Sets the value of :py:attr:`linkPredictionCol`. """ return self._set(linkPredictionCol=value)
def coefficients(self): """ Model coefficients. """
def intercept(self): """ Model intercept. """
def summary(self): """ Gets summary (residuals, deviance, p-values) of model on training set. An exception is thrown if `trainingSummary is None`. """ super(GeneralizedLinearRegressionModel, self).summary) else: raise RuntimeError("No training summary available for this %s" % self.__class__.__name__)
""" Evaluates the model on a test dataset.
.. versionadded:: 2.0.0
Parameters ---------- dataset : :py:class:`pyspark.sql.DataFrame` Test dataset to evaluate model on, where dataset is an instance of :py:class:`pyspark.sql.DataFrame` """
""" Generalized linear regression results evaluated on a dataset.
.. versionadded:: 2.0.0 """
def predictions(self): """ Predictions output by the model's `transform` method. """
def predictionCol(self): """ Field in :py:attr:`predictions` which gives the predicted value of each instance. This is set to a new column name if the original model's `predictionCol` is not set. """
def numInstances(self): """ Number of instances in DataFrame predictions. """
def rank(self): """ The numeric rank of the fitted linear model. """
def degreesOfFreedom(self): """ Degrees of freedom. """
def residualDegreeOfFreedom(self): """ The residual degrees of freedom. """
def residualDegreeOfFreedomNull(self): """ The residual degrees of freedom for the null model. """
""" Get the residuals of the fitted model by type.
.. versionadded:: 2.0.0
Parameters ---------- residualsType : str, optional The type of residuals which should be returned. Supported options: deviance (default), pearson, working, and response. """
def nullDeviance(self): """ The deviance for the null model. """
def deviance(self): """ The deviance for the fitted model. """
def dispersion(self): """ The dispersion of the fitted model. It is taken as 1.0 for the "binomial" and "poisson" families, and otherwise estimated by the residual Pearson's Chi-Squared statistic (which is defined as sum of the squares of the Pearson residuals) divided by the residual degrees of freedom. """
def aic(self): """ Akaike's "An Information Criterion"(AIC) for the fitted model. """
""" Generalized linear regression training results.
.. versionadded:: 2.0.0 """
def numIterations(self): """ Number of training iterations. """
def solver(self): """ The numeric solver used for training. """
def coefficientStandardErrors(self): """ Standard error of estimated coefficients and intercept.
If :py:attr:`GeneralizedLinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept. """
def tValues(self): """ T-statistic of estimated coefficients and intercept.
If :py:attr:`GeneralizedLinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept. """
def pValues(self): """ Two-sided p-value of estimated coefficients and intercept.
If :py:attr:`GeneralizedLinearRegression.fitIntercept` is set to True, then the last element returned corresponds to the intercept. """
return self._call_java("toString")
HasSolver, HasSeed, HasFitIntercept, HasRegParam, HasWeightCol): """ Params for :py:class:`FMRegressor`, :py:class:`FMRegressionModel`, :py:class:`FMClassifier` and :py:class:`FMClassifierModel`.
.. versionadded:: 3.0.0 """
"which are used to get pairwise interactions between variables", typeConverter=TypeConverters.toInt)
typeConverter=TypeConverters.toBoolean)
"set that should be used for one iteration of gradient descent", typeConverter=TypeConverters.toFloat)
typeConverter=TypeConverters.toFloat)
"options: gd, adamW. (Default adamW)", typeConverter=TypeConverters.toString)
miniBatchFraction=1.0, initStd=0.01, maxIter=100, stepSize=1.0, tol=1e-6, solver="adamW")
def getFactorSize(self): """ Gets the value of factorSize or its default value. """ return self.getOrDefault(self.factorSize)
def getFitLinear(self): """ Gets the value of fitLinear or its default value. """ return self.getOrDefault(self.fitLinear)
def getMiniBatchFraction(self): """ Gets the value of miniBatchFraction or its default value. """ return self.getOrDefault(self.miniBatchFraction)
def getInitStd(self): """ Gets the value of initStd or its default value. """ return self.getOrDefault(self.initStd)
""" Factorization Machines learning algorithm for regression.
solver Supports:
* gd (normal mini-batch gradient descent) * adamW (default)
.. versionadded:: 3.0.0
Examples -------- >>> from pyspark.ml.linalg import Vectors >>> from pyspark.ml.regression import FMRegressor >>> df = spark.createDataFrame([ ... (2.0, Vectors.dense(2.0)), ... (1.0, Vectors.dense(1.0)), ... (0.0, Vectors.sparse(1, [], []))], ["label", "features"]) >>> >>> fm = FMRegressor(factorSize=2) >>> fm.setSeed(16) FMRegressor... >>> model = fm.fit(df) >>> model.getMaxIter() 100 >>> test0 = spark.createDataFrame([ ... (Vectors.dense(-2.0),), ... (Vectors.dense(0.5),), ... (Vectors.dense(1.0),), ... (Vectors.dense(4.0),)], ["features"]) >>> model.transform(test0).show(10, False) +--------+-------------------+ |features|prediction | +--------+-------------------+ |[-2.0] |-1.9989237712341565| |[0.5] |0.4956682219523814 | |[1.0] |0.994586620589689 | |[4.0] |3.9880970124135344 | +--------+-------------------+ ... >>> model.intercept -0.0032501766849261557 >>> model.linear DenseVector([0.9978]) >>> model.factors DenseMatrix(1, 2, [0.0173, 0.0021], 1) >>> model_path = temp_path + "/fm_model" >>> model.save(model_path) >>> model2 = FMRegressionModel.load(model_path) >>> model2.intercept -0.0032501766849261557 >>> model2.linear DenseVector([0.9978]) >>> model2.factors DenseMatrix(1, 2, [0.0173, 0.0021], 1) >>> model.transform(test0).take(1) == model2.transform(test0).take(1) True """
factorSize=8, fitIntercept=True, fitLinear=True, regParam=0.0, miniBatchFraction=1.0, initStd=0.01, maxIter=100, stepSize=1.0, tol=1e-6, solver="adamW", seed=None): """ __init__(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ factorSize=8, fitIntercept=True, fitLinear=True, regParam=0.0, \ miniBatchFraction=1.0, initStd=0.01, maxIter=100, stepSize=1.0, \ tol=1e-6, solver="adamW", seed=None) """ "org.apache.spark.ml.regression.FMRegressor", self.uid)
factorSize=8, fitIntercept=True, fitLinear=True, regParam=0.0, miniBatchFraction=1.0, initStd=0.01, maxIter=100, stepSize=1.0, tol=1e-6, solver="adamW", seed=None): """ setParams(self, \\*, featuresCol="features", labelCol="label", predictionCol="prediction", \ factorSize=8, fitIntercept=True, fitLinear=True, regParam=0.0, \ miniBatchFraction=1.0, initStd=0.01, maxIter=100, stepSize=1.0, \ tol=1e-6, solver="adamW", seed=None) Sets Params for FMRegressor. """
def setFactorSize(self, value): """ Sets the value of :py:attr:`factorSize`. """ return self._set(factorSize=value)
def setFitLinear(self, value): """ Sets the value of :py:attr:`fitLinear`. """ return self._set(fitLinear=value)
def setMiniBatchFraction(self, value): """ Sets the value of :py:attr:`miniBatchFraction`. """ return self._set(miniBatchFraction=value)
def setInitStd(self, value): """ Sets the value of :py:attr:`initStd`. """ return self._set(initStd=value)
def setMaxIter(self, value): """ Sets the value of :py:attr:`maxIter`. """ return self._set(maxIter=value)
def setStepSize(self, value): """ Sets the value of :py:attr:`stepSize`. """ return self._set(stepSize=value)
def setTol(self, value): """ Sets the value of :py:attr:`tol`. """ return self._set(tol=value)
def setSolver(self, value): """ Sets the value of :py:attr:`solver`. """ return self._set(solver=value)
def setSeed(self, value): """ Sets the value of :py:attr:`seed`. """
def setFitIntercept(self, value): """ Sets the value of :py:attr:`fitIntercept`. """ return self._set(fitIntercept=value)
def setRegParam(self, value): """ Sets the value of :py:attr:`regParam`. """ return self._set(regParam=value)
JavaMLReadable): """ Model fitted by :class:`FMRegressor`.
.. versionadded:: 3.0.0 """
def intercept(self): """ Model intercept. """
def linear(self): """ Model linear term. """
def factors(self): """ Model factor term. """
# The small batch size here ensures that we see multiple batches, # even in these small test examples: .master("local[2]")\ .appName("ml.regression tests")\ .getOrCreate() finally: except OSError: pass sys.exit(-1) |