def convolve(f, g, mode='full', axes=None):
'''Perform a convolution along given axes'''
if mode == 'same':
return _signal.convolveToSameShape(_asf(f)._jdataset(), _asf(g)._jdataset(), axes)
elif mode == 'valid':
return _signal.convolveForOverlap(_asf(f)._jdataset(), _asf(g)._jdataset(), axes)
elif mode == 'full':
return _signal.convolve(_asf(f)._jdataset(), _asf(g)._jdataset(), axes)
raise ValueError, 'mode keyword has unrecognised value'
def phasecorrelate(f, g, axes=None, includeinv=False):
'''Perform a phase cross correlation along given axes (can include inverse of cross-power spectrum)'''
ans = _signal.phaseCorrelate(_asf(f)._jdataset(), _asf(g)._jdataset(), axes, includeinv)
if includeinv:
return ans[0], ans[1]
else:
return ans[0]
def correlate(f, g=None, mode='valid', old_behavior=False, axes=None):
'''Perform a cross (or auto if g is None) correlation along given axes'''
if old_behavior:
raise NotImplementedError, 'Not implemented'
f = _asf(f)._jdataset()
if g is None:
g = f
else:
g = _asf(g)._jdataset()
if mode == 'same':
return _signal.correlateToSameShape(f, g, axes)
elif mode == 'valid':
return _signal.correlateForOverlap(f, g, axes)
elif mode == 'full':
return _signal.correlate(f, g, axes)
raise ValueError, 'mode keyword has unrecognised value'
def convolve(f, g, axes=None):
'''Perform a convolution along given axes'''
return _signal.convolve(f, g, axes)
def correlate(f, g=None, axes=None):
'''Perform a cross (or auto if g is None) correlation along given axes'''
if g is None:
return _signal.correlate(f, axes)
return _signal.correlate(f, g, axes)
