apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-02 14:45:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Modified gradient_checker to allow for variable 'f'

Browse source
This commit is contained in:
Alan Saul 2013-09-11 11:54:47 +01:00
parent 1dd83291fe
commit 64e65b846d
1 changed files with 15 additions and 15 deletions
Download patch file Download diff file Expand all files Collapse all files

30
GPy/models/gradient_checker.py
View file

@ -26,40 +26,40 @@ class GradientChecker(Model):
""" """
:param f: Function to check gradient for :param f: Function to check gradient for
:param df: Gradient of function to check :param df: Gradient of function to check
:param x0: :param x0:
Initial guess for inputs x (if it has a shape (a,b) this will be reflected in the parameter names). Initial guess for inputs x (if it has a shape (a,b) this will be reflected in the parameter names).
Can be a list of arrays, if takes a list of arrays. This list will be passed Can be a list of arrays, if takes a list of arrays. This list will be passed
to f and df in the same order as given here. to f and df in the same order as given here.
If only one argument, make sure not to pass a list!!! If only one argument, make sure not to pass a list!!!
:type x0: [array-like] | array-like | float | int :type x0: [array-like] | array-like | float | int
:param names: :param names:
Names to print, when performing gradcheck. If a list was passed to x0 Names to print, when performing gradcheck. If a list was passed to x0
a list of names with the same length is expected. a list of names with the same length is expected.
:param args: Arguments passed as f(x, *args, **kwargs) and df(x, *args, **kwargs) :param args: Arguments passed as f(x, *args, **kwargs) and df(x, *args, **kwargs)
Examples: Examples:
--------- ---------
from GPy.models import GradientChecker from GPy.models import GradientChecker
N, M, Q = 10, 5, 3 N, M, Q = 10, 5, 3
Sinusoid: Sinusoid:
X = numpy.random.rand(N, Q) X = numpy.random.rand(N, Q)
grad = GradientChecker(numpy.sin,numpy.cos,X,'x') grad = GradientChecker(numpy.sin,numpy.cos,X,'x')
grad.checkgrad(verbose=1) grad.checkgrad(verbose=1)
Using GPy: Using GPy:
X, Z = numpy.random.randn(N,Q), numpy.random.randn(M,Q) X, Z = numpy.random.randn(N,Q), numpy.random.randn(M,Q)
kern = GPy.kern.linear(Q, ARD=True) + GPy.kern.rbf(Q, ARD=True) kern = GPy.kern.linear(Q, ARD=True) + GPy.kern.rbf(Q, ARD=True)
grad = GradientChecker(kern.K, grad = GradientChecker(kern.K,
lambda x: 2*kern.dK_dX(numpy.ones((1,1)), x), lambda x: 2*kern.dK_dX(numpy.ones((1,1)), x),
x0 = X.copy(), x0 = X.copy(),
names='X') names='X')
grad.checkgrad(verbose=1) grad.checkgrad(verbose=1)
grad.randomize() grad.randomize()
grad.checkgrad(verbose=1) grad.checkgrad(verbose=1)
""" """
Model.__init__(self) Model.__init__(self)
if isinstance(x0, (list, tuple)) and names is None: if isinstance(x0, (list, tuple)) and names is None:
@ -81,8 +81,8 @@ class GradientChecker(Model):
# self._param_names.extend(map(lambda nameshape: ('_'.join(nameshape)).strip('_'), itertools.izip(itertools.repeat(name), itertools.imap(lambda t: '_'.join(map(str, t)), itertools.product(*map(lambda xi: range(xi), shape)))))) # self._param_names.extend(map(lambda nameshape: ('_'.join(nameshape)).strip('_'), itertools.izip(itertools.repeat(name), itertools.imap(lambda t: '_'.join(map(str, t)), itertools.product(*map(lambda xi: range(xi), shape))))))
self.args = args self.args = args
self.kwargs = kwargs self.kwargs = kwargs
self.f = f self._f = f
self.df = df self._df = df
def _get_x(self): def _get_x(self):
if len(self.names) > 1: if len(self.names) > 1:
@ -90,10 +90,10 @@ class GradientChecker(Model):
return [self.__getattribute__(self.names[0])] + list(self.args) return [self.__getattribute__(self.names[0])] + list(self.args)
def log_likelihood(self): def log_likelihood(self):
return float(numpy.sum(self.f(*self._get_x(), **self.kwargs))) return float(numpy.sum(self._f(*self._get_x(), **self.kwargs)))
def _log_likelihood_gradients(self): def _log_likelihood_gradients(self):
return numpy.atleast_1d(self.df(*self._get_x(), **self.kwargs)).flatten() return numpy.atleast_1d(self._df(*self._get_x(), **self.kwargs)).flatten()
def _get_params(self): def _get_params(self):