Package random_distributions

This module provides functionality to generate random numbers with arbitrary distributions. More...

Classes
class	XYResponseCurve
	Similar to ResponseCurve, but allows sample points to be unevenly spaced. More...
class	ResponseCurve
	This class is described in AI Programming Wisdom 1, "The Beauty of Response Curves", by Bob Alexander. More...
class	NormalisedInputCurve
class	Distribution2D
	This is a callable class that generates random numbers with an arbitrary distribution. More...
Functions
def	lerp
	Linear interpolation, with clamping for inputs outside the range.
def	sigmoid
	This function is a "soft" version of lerp.
def	ramp
	The same as lerp, except that the output is extrapolated, not clamped when the input is higher than the maximum.
def	line
	The same as lerp, except that output is linearly extrapolated when the input falls outside the range on both ends.
def	make_distribution_curve
	Makes a distribution curve from input and output samples.
def	distribution_from_grid
Variables
int	newSampleCount = 1000
float	THRESHOLD = 0.001

---

Detailed Description

This module provides functionality to generate random numbers with arbitrary distributions.

See:

• http://code-spot.co.za/2008/09/21/generating-random-numbers-with-arbitrary-distributions/
• http://code-spot.co.za/2009/04/15/generating-random-points-from-arbitrary-distributions-for-2d-and-up/
• http://code-spot.co.za/2009/04/15/estimating-a-continuous-distribution-from-a-sample-set/

Author:

Herman Tulleken (herman.tulleken@gmail.com)

---

Function Documentation

def random_distributions.distribution_from_grid	(		grid,
			x_cells,
			y_cells
	)

Definition at line 308 of file random_distributions.py.

def random_distributions.lerp	(		value,
			inputMin,
			inputMax,
			outputMin,
			outputMax
	)

Linear interpolation, with clamping for inputs outside the range.

This function can also be used for linear extrapolation. Compare with ramp and

Definition at line 29 of file random_distributions.py.

def random_distributions.line	(		value,
			inputMin,
			inputMax,
			outputMin,
			outputMax
	)

The same as lerp, except that output is linearly extrapolated when the input falls outside the range on both ends.

Definition at line 52 of file random_distributions.py.

def random_distributions.make_distribution_curve	(		inputSamples,
			outputSamples
	)

Makes a distribution curve from input and output samples.

The inputSamples denote boundaries, and the outputSamples denote the relative probabilities that a point is between those boundaries. Let x_0..x_{n-1} denote the inputs, and y_0..y_{n-1} be the outputs. Then y_i gives the relative probability that a point will be selected from the range [x_{i-1}, x_i). y_0 should be 0.

Definition at line 216 of file random_distributions.py.

def random_distributions.ramp	(		value,
			inputMin,
			inputMax,
			outputMin,
			outputMax
	)

The same as lerp, except that the output is extrapolated, not clamped when the input is higher than the maximum.

The output is still clamped when the input fals below the inputMin.

Definition at line 44 of file random_distributions.py.

def random_distributions.sigmoid	(		value,
			inputMin,
			inputMax,
			outputMin,
			outputMax
	)

This function is a "soft" version of lerp.

Definition at line 36 of file random_distributions.py.

---

Variable Documentation

int random_distributions.newSampleCount = 1000

Definition at line 23 of file random_distributions.py.

float random_distributions.THRESHOLD = 0.001

Definition at line 24 of file random_distributions.py.