def test_gradient(self):
mcgsm = MCGSM(5, 2, 2, 4, 10)
cholesky_factors = []
for k in range(mcgsm.num_components):
cholesky_factors.append(cholesky(cov(randn(mcgsm.dim_out, mcgsm.dim_out ** 2))))
mcgsm.cholesky_factors = cholesky_factors
mcgsm.linear_features = randn(mcgsm.num_components, mcgsm.dim_in) / 5.0
mcgsm.means = randn(mcgsm.dim_out, mcgsm.num_components) / 5.0
err = mcgsm._check_gradient(randn(mcgsm.dim_in, 1000), randn(mcgsm.dim_out, 1000), 1e-5)
self.assertLess(err, 1e-8)
# without regularization
for param in ["priors", "scales", "weights", "features", "chol", "pred", "linear_features", "means"]:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-5,
parameters={
"train_prior": param == "priors",
"train_scales": param == "scales",
"train_weights": param == "weights",
"train_features": param == "features",
"train_cholesky_factors": param == "chol",
"train_predictors": param == "pred",
"train_linear_features": param == "linear_features",
"train_means": param == "means",
},
)
self.assertLess(err, 1e-8)
# with regularization
for norm in ["L1", "L2"]:
for param in ["priors", "scales", "weights", "features", "chol", "pred", "linear_features", "means"]:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-7,
parameters={
"train_prior": param == "priors",
"train_scales": param == "scales",
"train_weights": param == "weights",
"train_features": param == "features",
"train_cholesky_factors": param == "chol",
"train_predictors": param == "pred",
"train_linear_features": param == "linear_features",
"train_means": param == "means",
"regularize_features": {"strength": 0.4, "norm": norm},
"regularize_predictors": {"strength": 0.5, "norm": norm},
"regularize_weights": {"strength": 0.7, "norm": norm},
"regularize_linear_features": {"strength": 0.3, "norm": norm},
"regularize_means": {"strength": 0.6, "norm": norm},
},
)
self.assertLess(err, 1e-6)
def test_gradient(self):
mcgsm = MCGSM(5, 2, 2, 4, 10)
cholesky_factors = []
for k in range(mcgsm.num_components):
cholesky_factors.append(cholesky(cov(randn(mcgsm.dim_out, mcgsm.dim_out**2))))
mcgsm.cholesky_factors = cholesky_factors
mcgsm.linear_features = randn(mcgsm.num_components, mcgsm.dim_in) / 5.
mcgsm.means = randn(mcgsm.dim_out, mcgsm.num_components) / 5.
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000), 1e-5)
self.assertLess(err, 1e-8)
# without regularization
for param in ['priors', 'scales', 'weights', 'features', 'chol', 'pred', 'linear_features', 'means']:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-5,
parameters={
'train_prior': param == 'priors',
'train_scales': param == 'scales',
'train_weights': param == 'weights',
'train_features': param == 'features',
'train_cholesky_factors': param == 'chol',
'train_predictors': param == 'pred',
'train_linear_features': param == 'linear_features',
'train_means': param == 'means',
})
self.assertLess(err, 1e-8)
# with regularization
for norm in ['L1', 'L2']:
for param in ['priors', 'scales', 'weights', 'features', 'chol', 'pred', 'linear_features', 'means']:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-7,
parameters={
'train_prior': param == 'priors',
'train_scales': param == 'scales',
'train_weights': param == 'weights',
'train_features': param == 'features',
'train_cholesky_factors': param == 'chol',
'train_predictors': param == 'pred',
'train_linear_features': param == 'linear_features',
'train_means': param == 'means',
'regularize_features': {'strength': 0.4, 'norm': norm},
'regularize_predictors': {'strength': 0.5, 'norm': norm},
'regularize_weights': {'strength': 0.7, 'norm': norm},
'regularize_linear_features': {'strength': 0.3, 'norm': norm},
'regularize_means': {'strength': 0.6, 'norm': norm},
})
self.assertLess(err, 1e-6)
def test_data_gradient(self):
for dim_in in [5, 0]:
mcgsm = MCGSM(dim_in, 3, 4, 5, 10)
cholesky_factors = []
for k in range(mcgsm.num_components):
cholesky_factors.append(
cholesky(cov(randn(mcgsm.dim_out, mcgsm.dim_out**2))))
mcgsm.cholesky_factors = cholesky_factors
inputs = randn(mcgsm.dim_in, 100)
outputs = ones_like(mcgsm.sample(inputs))
# compute density gradient and loglikelihood
dx, dy, ll = mcgsm._data_gradient(inputs, outputs)
self.assertLess(
max(abs(ll - mcgsm.loglikelihood(inputs, outputs))), 1e-8)
h = 1e-5
dx_ = zeros_like(dx)
dy_ = zeros_like(dy)
for i in range(mcgsm.dim_in):
inputs_p = inputs.copy()
inputs_m = inputs.copy()
inputs_p[i] += h
inputs_m[i] -= h
dx_[i] = (mcgsm.loglikelihood(inputs_p, outputs) -
mcgsm.loglikelihood(inputs_m, outputs)) / (2. * h)
for i in range(mcgsm.dim_out):
outputs_p = outputs.copy()
outputs_m = outputs.copy()
outputs_p[i] += h
outputs_m[i] -= h
dy_[i] = (mcgsm.loglikelihood(inputs, outputs_p) -
mcgsm.loglikelihood(inputs, outputs_m)) / (2. * h)
self.assertLess(max(abs(dy_ - dy)), 1e-8)
if mcgsm.dim_in > 0:
self.assertLess(max(abs(dx_ - dx)), 1e-8)
def test_data_gradient(self): for dim_in in [5, 0]: mcgsm = MCGSM(dim_in, 3, 4, 5, 10) cholesky_factors = [] for k in range(mcgsm.num_components): cholesky_factors.append(cholesky(cov(randn(mcgsm.dim_out, mcgsm.dim_out**2)))) mcgsm.cholesky_factors = cholesky_factors inputs = randn(mcgsm.dim_in, 100) outputs = ones_like(mcgsm.sample(inputs)) # compute density gradient and loglikelihood dx, dy, ll = mcgsm._data_gradient(inputs, outputs) self.assertLess(max(abs(ll - mcgsm.loglikelihood(inputs, outputs))), 1e-8) h = 1e-5 dx_ = zeros_like(dx) dy_ = zeros_like(dy) for i in range(mcgsm.dim_in): inputs_p = inputs.copy() inputs_m = inputs.copy() inputs_p[i] += h inputs_m[i] -= h dx_[i] = ( mcgsm.loglikelihood(inputs_p, outputs) - mcgsm.loglikelihood(inputs_m, outputs)) / (2. * h) for i in range(mcgsm.dim_out): outputs_p = outputs.copy() outputs_m = outputs.copy() outputs_p[i] += h outputs_m[i] -= h dy_[i] = ( mcgsm.loglikelihood(inputs, outputs_p) - mcgsm.loglikelihood(inputs, outputs_m)) / (2. * h) self.assertLess(max(abs(dy_ - dy)), 1e-8) if mcgsm.dim_in > 0: self.assertLess(max(abs(dx_ - dx)), 1e-8)
def test_gradient(self):
mcgsm = MCGSM(5, 2, 2, 4, 10)
cholesky_factors = []
for k in range(mcgsm.num_components):
cholesky_factors.append(
cholesky(cov(randn(mcgsm.dim_out, mcgsm.dim_out**2))))
mcgsm.cholesky_factors = cholesky_factors
mcgsm.linear_features = randn(mcgsm.num_components, mcgsm.dim_in) / 5.
mcgsm.means = randn(mcgsm.dim_out, mcgsm.num_components) / 5.
err = mcgsm._check_gradient(randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000), 1e-5)
self.assertLess(err, 1e-8)
# without regularization
for param in [
'priors', 'scales', 'weights', 'features', 'chol', 'pred',
'linear_features', 'means'
]:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-5,
parameters={
'train_prior': param == 'priors',
'train_scales': param == 'scales',
'train_weights': param == 'weights',
'train_features': param == 'features',
'train_cholesky_factors': param == 'chol',
'train_predictors': param == 'pred',
'train_linear_features': param == 'linear_features',
'train_means': param == 'means',
})
self.assertLess(err, 1e-8)
# with regularization
for norm in ['L1', 'L2']:
for param in [
'priors', 'scales', 'weights', 'features', 'chol', 'pred',
'linear_features', 'means'
]:
err = mcgsm._check_gradient(
randn(mcgsm.dim_in, 1000),
randn(mcgsm.dim_out, 1000),
1e-7,
parameters={
'train_prior': param == 'priors',
'train_scales': param == 'scales',
'train_weights': param == 'weights',
'train_features': param == 'features',
'train_cholesky_factors': param == 'chol',
'train_predictors': param == 'pred',
'train_linear_features': param == 'linear_features',
'train_means': param == 'means',
'regularize_features': {
'strength': 0.4,
'norm': norm
},
'regularize_predictors': {
'strength': 0.5,
'norm': norm
},
'regularize_weights': {
'strength': 0.7,
'norm': norm
},
'regularize_linear_features': {
'strength': 0.3,
'norm': norm
},
'regularize_means': {
'strength': 0.6,
'norm': norm
},
})
self.assertLess(err, 1e-6)
