Table Of Contents

Previous topic

io.pyniftiio

Next topic

modalities.fmri.fmri

This Page

modalities.fmri.filters

Module: modalities.fmri.filters

Inheritance diagram for nipy.modalities.fmri.filters:

TODO

Classes

FIR

class nipy.modalities.fmri.filters.FIR(parameters)

Bases: nipy.modalities.fmri.filters.Filter

A class for FIR filters: i.e. the filter is a collection of square waves. Parameters (start and duration) are specified as a kx2 matrix for k square waves.

>>> GUI = True
>>> from nipy.modalities.fmri import filters
>>> from pylab import *
>>> from numpy import *
>>> parameters = array([[1., 2.], [2., 5.], [4., 8.]])
>>> IRF = filters.FIR(parameters)
>>> _ = plot(arange(0, 15, 0.1), sum(IRF(arange(0, 15, 0.1)), axis=0))
>>> ylab = ylabel('Filters')
>>> xlab = xlabel('Time (s)')
>>> show()
__init__(parameters)
Parameters:
parameters : TODO

TODO

Filter

class nipy.modalities.fmri.filters.Filter(IRF, names, dt=0.02, tmin=-10.0, tmax=500.0)

Bases: object

Takes a list of impulse response functions (IRFs): main purpose is to convolve a functions with each IRF for Design. The class assumes the range of the filter is effectively 50 seconds, can be changed by setting tmax – this is just for the __mul__ method for convolution.

__init__(IRF, names, dt=0.02, tmin=-10.0, tmax=500.0)
Parameters:
IRF : TODO

TODO

names : TODO

TODO

dt : float

TODO

tmin : float

TODO

tmax : float

TODO

convolve(fn, interval=None, dt=None)
Take a (possibly vector-valued) function fn of time and return a linearly interpolated function after convolving with the filter.

GammaCOMB

class nipy.modalities.fmri.filters.GammaCOMB(fns)

TODO

__init__(fns)
Parameters:
fns : TODO

TODO

deriv(const=1.0)
Parameters:
const : float

TODO

Returns:

GammaCOMB

GammaDENS

class nipy.modalities.fmri.filters.GammaDENS(alpha, nu, coef=1.0)

A class for a Gamma density which knows how to differentiate itself.

By default, normalized to integrate to 1.

__init__(alpha, nu, coef=1.0)
Parameters:
alpha : TODO

TODO

nu : TODO

TODO

coef : float

TODO

deriv(const=1.0)
Differentiate a Gamma density. Returns a GammaCOMB that can evaluate the derivative.

GammaHRF

class nipy.modalities.fmri.filters.GammaHRF(parameters)

Bases: nipy.modalities.fmri.filters.Filter

A class that represents the Gamma basis in SPM: i.e. the filter is a collection of a certain number of Gamma densities. Parameters are specified as a kx2 matrix for k Gamma functions.

__init__(parameters)
Parameters:
parameters : TODO

TODO

deriv(const=1.0)
Parameters:
const : float

TODO

Returns:

TODO