def infer_dims(self, x, y, dims_x, dims_y, dims_out):
"""Infer probable output from input x, y
"""
OptimizedInverseModel.infer_x(self, y)
assert len(x) == len(dims_x)
assert len(y) == len(dims_y)
if len(self.fmodel.dataset) == 0:
return [[0.0] * self.dim_out]
else:
_, index = self.fmodel.dataset.nn_dims(x, y, dims_x, dims_y, k=1)
guesses = [self.fmodel.dataset.get_dims(index[0], dims_out)]
result = []
for g in guesses:
res = cma.fmin(
lambda q: self._error_dims(q, dims_x, dims_y, dims_out),
g,
self.cmaes_sigma,
options={
'bounds': [self.lower, self.upper],
'verb_log': 0,
'verb_disp': False,
'maxfevals': self.maxfevals
})
result.append((res[1], res[0]))
return sorted(result)[0][1]
def infer_x(self, y):
"""Infer probable x from input y
@param y the desired output for infered x.
@return a list of probable x
"""
OptimizedInverseModel.infer_x(self, y)
if self.fmodel.size() == 0:
return self._random_x()
x_guesses = [self._guess_x_simple(y)[0]]
result = []
for xg in x_guesses:
res = cma.fmin(self._error,
xg,
self.cmaes_sigma,
options={
'bounds': [self.lower, self.upper],
'verb_log': 0,
'verb_disp': False,
'maxfevals': self.maxfevals
})
result.append((res[1], res[0]))
return [xi for fi, xi in sorted(result)]
def infer_x(self, y):
"""Infer probable x from input y
@param y the desired output for infered x.
@return a list of probable x
"""
OptimizedInverseModel.infer_x(self, y)
if self.fmodel.size() == 0:
return self._random_x()
x_guesses = [self._guess_x_simple(y)[0]]
result = []
for xg in x_guesses:
res = cma.fmin(self._error, xg, self.cmaes_sigma,
options={'bounds':[self.lower, self.upper],
'verb_log':0,
'verb_disp':False,
'maxfevals':self.maxfevals})
result.append((res[1], res[0]))
return [xi for fi, xi in sorted(result)]
def infer_dims(self, x, y, dims_x, dims_y, dims_out):
"""Infer probable output from input x, y
"""
OptimizedInverseModel.infer_x(self, y)
assert len(x) == len(dims_x)
assert len(y) == len(dims_y)
if len(self.fmodel.dataset) == 0:
return [[0.0]*self.dim_out]
else:
_, index = self.fmodel.dataset.nn_dims(x, y, dims_x, dims_y, k=1)
guesses = [self.fmodel.dataset.get_dims(index[0], dims_out)]
result = []
for g in guesses:
res = cma.fmin(lambda q:self._error_dims(q, dims_x, dims_y, dims_out), g, self.cmaes_sigma,
options={'bounds':[self.lower, self.upper],
'verb_log':0,
'verb_disp':False,
'maxfevals':self.maxfevals})
result.append((res[1], res[0]))
return sorted(result)[0][1]
