diff options
Diffstat (limited to 'tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py')
| -rw-r--r-- | tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py | 74 |
1 files changed, 47 insertions, 27 deletions
diff --git a/tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py b/tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py index a0c01ffc48..5a9b1603e7 100644 --- a/tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py +++ b/tensorflow/contrib/bayesflow/python/kernel_tests/variational_inference_test.py @@ -18,11 +18,28 @@ from __future__ import absolute_import from __future__ import division from __future__ import print_function -import tensorflow as tf - -st = tf.contrib.bayesflow.stochastic_tensor -vi = tf.contrib.bayesflow.variational_inference -distributions = tf.contrib.distributions +import sys + +# TODO: #6568 Remove this hack that makes dlopen() not crash. +if hasattr(sys, "getdlopenflags") and hasattr(sys, "setdlopenflags"): + import ctypes + sys.setdlopenflags(sys.getdlopenflags() | ctypes.RTLD_GLOBAL) + +from tensorflow.contrib import distributions as distributions_lib +from tensorflow.contrib import layers +from tensorflow.contrib.bayesflow.python.ops import stochastic_tensor +from tensorflow.contrib.bayesflow.python.ops import variational_inference +from tensorflow.contrib.distributions.python.ops import kullback_leibler +from tensorflow.contrib.distributions.python.ops import normal +from tensorflow.python.framework import constant_op +from tensorflow.python.ops import array_ops +from tensorflow.python.ops import math_ops +from tensorflow.python.ops import variables +from tensorflow.python.platform import test + +st = stochastic_tensor +vi = variational_inference +distributions = distributions_lib class NormalNoEntropy(distributions.Normal): @@ -33,45 +50,46 @@ class NormalNoEntropy(distributions.Normal): # For mini-VAE def inference_net(x, latent_size): - return tf.contrib.layers.linear(x, latent_size) + return layers.linear(x, latent_size) def generative_net(z, data_size): - return tf.contrib.layers.linear(z, data_size) + return layers.linear(z, data_size) def mini_vae(): x = [[-6., 3., 6.], [-8., 4., 8.]] prior = distributions.Normal(mu=0., sigma=1.) variational = st.StochasticTensor( - distributions.Normal(mu=inference_net(x, 1), sigma=1.)) + distributions.Normal( + mu=inference_net(x, 1), sigma=1.)) vi.register_prior(variational, prior) px = distributions.Normal(mu=generative_net(variational, 3), sigma=1.) - log_likelihood = tf.reduce_sum(px.log_prob(x), 1) - log_likelihood = tf.expand_dims(log_likelihood, -1) + log_likelihood = math_ops.reduce_sum(px.log_prob(x), 1) + log_likelihood = array_ops.expand_dims(log_likelihood, -1) return x, prior, variational, px, log_likelihood -class VariationalInferenceTest(tf.test.TestCase): +class VariationalInferenceTest(test.TestCase): def testDefaultVariationalAndPrior(self): _, prior, variational, _, log_likelihood = mini_vae() elbo = vi.elbo(log_likelihood) - expected_elbo = log_likelihood - tf.contrib.distributions.kl( + expected_elbo = log_likelihood - kullback_leibler.kl( variational.distribution, prior) with self.test_session() as sess: - sess.run(tf.global_variables_initializer()) + sess.run(variables.global_variables_initializer()) self.assertAllEqual(*sess.run([expected_elbo, elbo])) def testExplicitVariationalAndPrior(self): with self.test_session() as sess: _, _, variational, _, log_likelihood = mini_vae() - prior = tf.contrib.distributions.Normal(mu=3., sigma=2.) + prior = normal.Normal(mu=3., sigma=2.) elbo = vi.elbo( log_likelihood, variational_with_prior={variational: prior}) - expected_elbo = log_likelihood - tf.contrib.distributions.kl( + expected_elbo = log_likelihood - kullback_leibler.kl( variational.distribution, prior) - sess.run(tf.global_variables_initializer()) + sess.run(variables.global_variables_initializer()) self.assertAllEqual(*sess.run([expected_elbo, elbo])) def testExplicitForms(self): @@ -79,8 +97,9 @@ class VariationalInferenceTest(tf.test.TestCase): elbos = [] forms = vi.ELBOForms - for form in [forms.default, forms.analytic_kl, forms.sample, - forms.analytic_entropy]: + for form in [ + forms.default, forms.analytic_kl, forms.sample, forms.analytic_entropy + ]: elbo = vi.elbo( log_likelihood=log_likelihood, variational_with_prior={variational: prior}, @@ -88,23 +107,24 @@ class VariationalInferenceTest(tf.test.TestCase): elbos.append(elbo) with self.test_session() as sess: - sess.run(tf.global_variables_initializer()) - log_likelihood_shape = tf.shape(log_likelihood).eval() + sess.run(variables.global_variables_initializer()) + log_likelihood_shape = array_ops.shape(log_likelihood).eval() for elbo in elbos: elbo.eval() - elbo_shape = tf.shape(elbo).eval() + elbo_shape = array_ops.shape(elbo).eval() self.assertAllEqual(log_likelihood_shape, elbo_shape) self.assertEqual(elbo.dtype, log_likelihood.dtype) def testDefaultsSampleKLWithoutAnalyticKLOrEntropy(self): - x = tf.constant([[-6., 3., 6.]]) + x = constant_op.constant([[-6., 3., 6.]]) prior = distributions.Bernoulli(0.5) variational = st.StochasticTensor( - NormalNoEntropy(mu=inference_net(x, 1), sigma=1.)) + NormalNoEntropy( + mu=inference_net(x, 1), sigma=1.)) vi.register_prior(variational, prior) px = distributions.Normal(mu=generative_net(variational, 3), sigma=1.) - log_likelihood = tf.reduce_sum(px.log_prob(x), 1) + log_likelihood = math_ops.reduce_sum(px.log_prob(x), 1) # No analytic KL available between prior and variational distributions. with self.assertRaisesRegexp(NotImplementedError, "No KL"): @@ -117,7 +137,7 @@ class VariationalInferenceTest(tf.test.TestCase): variational) - variational.distribution.log_prob(variational) with self.test_session() as sess: - sess.run(tf.global_variables_initializer()) + sess.run(variables.global_variables_initializer()) self.assertAllEqual(*sess.run([expected_elbo, elbo])) def testElboWithLogJoint(self): @@ -125,9 +145,9 @@ class VariationalInferenceTest(tf.test.TestCase): _, prior, variational, _, log_likelihood = mini_vae() log_joint = log_likelihood + prior.log_prob(variational) elbo = vi.elbo_with_log_joint(log_joint) - sess.run(tf.global_variables_initializer()) + sess.run(variables.global_variables_initializer()) elbo.eval() if __name__ == "__main__": - tf.test.main() + test.main() |