A simple class to create and manipulate histograms. More...

#include <histogram.h>

Public Member Functions
	MyHistogram (unsigned int numBins=10, MyNT low=0, MyNT high=1)
virtual	~MyHistogram ()
virtual MyNT	computeMean ()
virtual MyNT	computeStandardDeviation (MyNT mean)
virtual MyNT	computeVariance (MyNT mean)
virtual unsigned int	getNumBins () const
virtual MyNT	getBin (MyNT coord) const
	Return the lower bound of the bin containing the argument.
virtual void	setBin (MyNT coord, MyNT value)
	Set the bin corresponding to coord to contain the argument value.
virtual MyNT	getValue (MyNT coord) const
	Return the value in the bin corresponding to coord.
MyNT	getMin () const
MyNT	getMax () const
MyNT	getRange () const
virtual void	insert (MyNT value)
virtual void	insert (const vector< MyNT > &values)
virtual void	insert (const vector< unsigned int > &values)
virtual void	abs ()
virtual void	accumulate (bool reverse=false)
void	constructTDistributionHistogram (unsigned int dof)
virtual MyNT	convertToPvalues (MyNT pvalueThreshold=0.01, LIBENRICHMENT_TEST_TYPE testType=LIBENRICHMENT_NONE, MyHistogram *reference=NULL)
	Convert the values in the histogram into p-values by normalising and computing the cumulative distribution.
virtual void	convertToPvalues (MyNT pvalueThreshold, const set< LIBENRICHMENT_TEST_TYPE > &testTypes, map< LIBENRICHMENT_TEST_TYPE, MyNT > &thresholds, map< LIBENRICHMENT_TEST_TYPE, MyHistogram > &histograms, MyHistogram *reference=NULL)
	Convert the values in the histogram into p-values by normalising and computing the cumulative distribution. Use each test in testTypes to correct for multiple hypotheses testing.
virtual const MyHistogram &	operator= (const MyHistogram &other)
virtual MyHistogram	operator+ (const MyHistogram &other)
virtual MyHistogram	operator- (MyNT weight)
virtual MyHistogram	operator- ()
virtual const MyHistogram &	operator+= (const MyHistogram &other)
virtual MyHistogram	operator* (MyNT weight)
	Return a new histogram where each bin is the product of each bin in this histogram and weight.
virtual void	operator*= (MyNT weight)
virtual MyHistogram	operator/ (MyHistogram &other)
virtual const MyHistogram &	operator/= (MyHistogram &other)
virtual void	normalise ()
virtual MyNT	getNumTotalValues () const
virtual void	print (ostream &ostr, string extra="", vector< MyHistogram > *others=NULL)
virtual void	print (string file, string extra="", vector< MyHistogram > *others=NULL)
virtual void	read (string infile)
virtual double	computeTrapezoidalAUC (const MyHistogram &other) const

Detailed Description

A simple class to create and manipulate histograms.

Author:: T. M. Murali (murali@cs.vt.edu). A class to create and manipulate histograms.

Constructor & Destructor Documentation

MyHistogram::MyHistogram	(	unsigned int	numBins = `10`,
		MyNT	low = `0`,
		MyNT	high = `1`
	)		`[inline]`

The constructor.

Parameters:

unsigned	int numBins, the number of bins in the histogram. The default value is 10.
MyNT	low, the smallest value to be stored in the histogram. The default value is 0.
MyNT	high, the largest value to be stored in the histogram. The default value is 1.

Remarks:: The class does not provide any other way to set the limits of the histogram, so as to prevent changing the limits after some values have been inserted into this histogram.

virtual MyHistogram::~MyHistogram ( ) [inline, virtual]

The destructor.

Member Function Documentation

void MyHistogram::abs ( ) [virtual]

Compute the absolute value of the histogram.

For every bin with a negative

void MyHistogram::accumulate ( bool reverse = false ) [virtual]

Set the value of each bin to be the partial sum of all the values in earlier bins.

Parameters:

[in] reverse,a boolean; if true, compute the partial sum in reverse.

This method is useful for converting a normalised histogram into a histogram of p-values, or with reverse set to true, into a histogram of probabilities.

MyNT MyHistogram::computeMean ( ) [virtual]

Compute the mean of the values stored in the histogram.

The method approximates each value stored by the lower endpoint of the bin the value falls in.

MyNT MyHistogram::computeStandardDeviation ( MyNT mean ) [virtual]

Compute the standard deviation of the values stored in the histogram.

Parameters:

in[	mean, the result of invoking MyHistogram::computeMean() on the invocant.

The method approximates each value stored by the lower endpoint of the bin the value falls in.

MyNT MyHistogram::computeVariance ( MyNT mean ) [virtual]

Compute the variance of the values stored in the histogram.

Parameters:

in[	mean, the result of invoking MyHistogram::computeMean() on the invocant.

The method approximates each value stored by the lower endpoint of the bin the value falls in.

void MyHistogram::constructTDistributionHistogram ( unsigned int dof )

Fill the histogram with values of the tail of Student's t distribution with dof degrees of freedom.

Parameters:

[in] dof,the number of degrees of freedom of the t distribution.

For each bin in the histogram, the method calculates the value of the (tail of the) t distribution at the left endpoint of the bin and stores that value in the bin.

MyNT MyHistogram::getMax ( ) const [inline]

Return the maximum value to be stored in the histogram.

MyNT MyHistogram::getMin ( ) const [inline]

Return the minimum value to be stored in the histogram.

virtual unsigned int MyHistogram::getNumBins ( ) const [inline, virtual]

Return the number of bins

virtual MyNT MyHistogram::getNumTotalValues ( ) const [inline, virtual]

Return the total number of values stored in the histogram.

MyNT MyHistogram::getRange ( ) const [inline]

Return the range of values to be stored in the histogram.

void MyHistogram::insert ( MyNT value ) [virtual]

Insert a value into to the histogram.

void MyHistogram::insert ( const vector< MyNT > & values ) [virtual]

Insert a vector of values into the histogram.

Insert a value into to the histogram. If value is less than min(), store value in the smallest bin. If value is larger than max(), store value in the largest bin.

Parameters:

MyNT	value, the number to be inserted.

void MyHistogram::normalise ( ) [virtual]

Ensure that the total number of values over all the bins is 1. The method simply divides the value in each bin by the total number of values in all the bins.

Note:: This method is equivalent to invoking MyHistogram::operator*() with the argument getNumTotalValues().

void MyHistogram::operator*= ( MyNT weight ) [virtual]

Multiply each bin in this histogram by weight.

Parameters:

[in] weight,the number to multiply each bin by.

MyHistogram MyHistogram::operator+ ( const MyHistogram & other ) [virtual]

Add two histograms.

Add other histogram to this histogram and return a new histogram containing the sum.

Warning:: Both histograms must have identical bins.

Parameters:

MyHistogram other, the other histogram to be added to the invocant.

Returns:: A MyHistogram containing the sum of the two histograms.

const MyHistogram & MyHistogram::operator+= ( const MyHistogram & other ) [virtual]

Increment invocant by another histogram.

Warning:: Both histograms must have identical bins.

Parameters:

MyHistogram other, the other histogram to be added to the invocant.

MyHistogram MyHistogram::operator- ( MyNT weight ) [virtual]

Subtract a number from the histogram.

MyHistogram MyHistogram::operator- ( ) [virtual]

Negate a number from the histogram.

MyHistogram MyHistogram::operator/ ( MyHistogram & other ) [virtual]

Create a new histogram by dividing the value in each bin in this histogram by the value in the bin in other.

Warning:: The behaviour of this method is undefined if the invocant and other do not have exactly the same range and number of bins.

const MyHistogram & MyHistogram::operator/= ( MyHistogram & other ) [virtual]

Dividing the value in each bin in this histogram by the value in the bin in other.

Warning:: The behaviour of this method is undefined if the invocant and other do not have exactly the same range and number of bins.

const MyHistogram & MyHistogram::operator= ( const MyHistogram & other ) [virtual]

Set the invocant equal to other.

Set invocant equal to another histogram.

Parameters:

MyHistogram other, the other histogram that the invocant will equal.

void MyHistogram::print	(	ostream &	ostr,
		string	extra = `""`,
		vector< MyHistogram > *	others = `NULL`
	)		`[virtual]`

Print the histogram.

Parameters:

[in]	ostr,a	reference to an output stream.
[in]	extra,an	"extra" string to print at the beginning of the output. Usually, this string is a comment. If you want to the output file to be processed by a plotting programme such as gnuplot, make sure that the extra string starts with a comment character, e.g., '#'.
[in]	otherHistograms,a	pointer to a vector of MyHistograms to print "in parallel" to the invocant.

Warning:: The method assumes that all histograms have the same limits and number of bins.

void MyHistogram::print	(	string	file,
		string	extra = `""`,
		vector< MyHistogram > *	others = `NULL`
	)		`[virtual]`

Print the histogram

Print the histogram to an output stream.

Parameters:

[in]	file,a	string specifying the file to print the histogram to.
[in]	extra,an	"extra" string to print at the beginning of the output. Usually, this string is a comment. If you want to the output file to be processed by a plotting programme such as gnuplot, make sure that the extra string starts with a comment character, e.g., '#'.
[in]	otherHistograms,a	pointer to a vector of MyHistograms to print "in parallel" to the invocant.

Warning:: The method assumes that all histograms have the same limits and number of bins.

void MyHistogram::read ( string infile ) [virtual]

Read the histogram from a file containing bin-value pairs.

The method assumes that blank lines separate different histograms.

Warning:: The behaviour of this method is undefined if infile contains multiple histograms with different ranges or differing number of bins.

The documentation for this class was generated from the following files:

/home/poirel/src/c++/biorithm/libutil/histogram.h
/home/poirel/src/c++/biorithm/libutil/histogram.C

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

Member Function Documentation