python.utils.pca_normalize_dataset Namespace Reference

Functions
def	clean_data (data)
	leave only the data points with coordinated greater than zero is convinient for some problems, i.e. More...
def	create_logger ()
def	norm (mat)
	normalization on [0,1] interval More...
def	remove_outliers (mat, koef, target=None)
	remove outliers, where the points deviate on more than koef times standard deviation from the mean More...

Variables
	action
	args
	C = cov(interest_data)
	data = genfromtxt(options.csv_dir + filename, skiprows=1, delimiter=',')
	default
	delimiter
	dest
	filename = options.file_in
	help
	interest_data = data - means
	interest_data_transformed = dot(invV, interest_data).T
	invV = inv(V)
def	logger = create_logger()
	means = mean(data,axis=0)
	options
	parser = optparse.OptionParser()
	target = data[:, options.target_column]
	type
	u
	V

Function Documentation

clean_data()

def python.utils.pca_normalize_dataset.clean_data

(

data

)

leave only the data points with coordinated greater than zero is convinient for some problems, i.e.

photometric redshift

create_logger()

def python.utils.pca_normalize_dataset.create_logger

(

)

norm()

def python.utils.pca_normalize_dataset.norm

(

mat

)

normalization on [0,1] interval

Parameters

mat	matrix points row-wise

remove_outliers()

def python.utils.pca_normalize_dataset.remove_outliers	(		mat,
			koef,
			target = None
	)

remove outliers, where the points deviate on more than koef times standard deviation from the mean

Parameters

mat	matrix points row-wise
koef	koefficient to determine the outliers
target	list with target values for the points

References python.datasetAnalysis.mean.

Variable Documentation

action

python.utils.pca_normalize_dataset.action

args

python.utils.pca_normalize_dataset.args

C

python.utils.pca_normalize_dataset.C = cov(interest_data)

Referenced by sgpp::datadriven::AlgorithmAdaBoostIdentity.alphaSolver(), sgpp::datadriven::SparseGridDensityEstimator.computeDensitySystemMatrix(), sgpp::datadriven::SparseGridDensityEstimator.computeRegularizationMatrix(), sgpp::datadriven::LearnerSGDE.computeRegularizationMatrix(), sgpp::datadriven::LearnerSGDE.computeResidual(), sgpp::datadriven::Learner.createDMSystem(), sgpp::optimization::optimizer::CMAES.optimize(), sgpp::datadriven::DBMatOnlineDE.resDensity(), sgpp::combigrid::BSplineInterpolationCoefficientEvaluator.setGridPoints(), sgpp::datadriven::LearnerSGDE.train(), sgpp::datadriven::LearnerSGDE.trainOnline(), and sgpp::datadriven::ModelFittingDensityEstimationCG.update().

data

def python.utils.pca_normalize_dataset.data = genfromtxt(options.csv_dir + filename, skiprows=1, delimiter=',')

default

python.utils.pca_normalize_dataset.default

delimiter

python.utils.pca_normalize_dataset.delimiter

dest

python.utils.pca_normalize_dataset.dest

filename

python.utils.pca_normalize_dataset.filename = options.file_in

help

python.utils.pca_normalize_dataset.help

interest_data

python.utils.pca_normalize_dataset.interest_data = data - means

interest_data_transformed

def python.utils.pca_normalize_dataset.interest_data_transformed = dot(invV, interest_data).T

invV

python.utils.pca_normalize_dataset.invV = inv(V)

logger

def python.utils.pca_normalize_dataset.logger = create_logger()

means

python.utils.pca_normalize_dataset.means = mean(data,axis=0)

Referenced by sgpp::datadriven::KernelDensityEstimator.cov(), and sgpp::datadriven::OperationCovariance.doQuadrature().

options

python.utils.pca_normalize_dataset.options

parser

python.utils.pca_normalize_dataset.parser = optparse.OptionParser()

target

python.utils.pca_normalize_dataset.target = data[:, options.target_column]

type

python.utils.pca_normalize_dataset.type

u

python.utils.pca_normalize_dataset.u

Referenced by sgpp::datadriven::DBMatDMSDenseIChol.choleskyBackwardSolve(), sgpp::datadriven::DBMatDMSDenseIChol.choleskyForwardSolve(), sgpp::datadriven::DBMatOfflineDenseIChol.choleskyModification(), sgpp::datadriven::DBMatOfflineDenseIChol.decomposeMatrix(), python.uq.estimators.MarginalIntegralStrategy.MarginalIntegralStrategy.estimate(), sgpp::optimization.fastPow(), sgpp::datadriven::ModelFittingClassification.fit(), sgpp::datadriven::DBMatOfflineDenseIChol.ichol(), sgpp::base::ConvertLinearToPrewavelet.operator()(), and sgpp::datadriven::LearnerSGDEOnOffParallel.predict().

V

python.utils.pca_normalize_dataset.V

Referenced by sgpp::combigrid::CombigridEvaluator< V >.addLevel(), sgpp::combigrid::AbstractLinearEvaluator< FloatTensorVector >.clone(), sgpp::combigrid::FullGridLinearSummationStrategy< V >.eval(), sgpp::combigrid::FullGridPCESummationStrategy< V >.eval(), sgpp::combigrid::FullGridTensorVarianceSummationStrategy< V >.eval(), sgpp::combigrid::FullGridOptimizedPCESummationStrategy< V >.eval(), sgpp::combigrid::AbstractLinearEvaluator< FloatTensorVector >.eval(), sgpp::combigrid::FullGridQuadraticSummationStrategy< V >.eval(), sgpp::combigrid::FullGridVarianceSummationStrategy< V >.eval(), sgpp::optimization::math.hessenbergForm(), sgpp::optimization::math.schurDecomposition(), sgpp::combigrid::AbstractEvaluator< FloatTensorVector >.setLevel(), sgpp::combigrid::AbstractFullGridEvaluator< V >.setMutex(), and sgpp::combigrid::AbstractFullGridSummationStrategy< V >.~AbstractFullGridSummationStrategy().