| /external/tensorflow/tensorflow/python/distribute/ |
| D | mirrored_strategy_test.py | 67 distribution=[
78 def testMinimizeLoss(self, distribution): argument
80 self._test_minimize_loss_eager(distribution)
82 self._test_minimize_loss_graph(distribution)
84 def testReplicaId(self, distribution): argument
85 self._test_replica_id(distribution)
87 def testNumReplicasInSync(self, distribution): argument
88 self.assertEqual(2, distribution.num_replicas_in_sync)
90 def testCallAndMergeExceptions(self, distribution): argument
91 self._test_call_and_merge_exceptions(distribution)
[all …]
|
| D | custom_training_loop_input_test.py | 79 distribution=strategy_combinations.all_strategies,
82 def testConstantNumpyInput(self, distribution): argument
90 outputs = distribution.experimental_local_results(
91 distribution.run(computation, args=(x,)))
95 constant_op.constant(4., shape=(distribution.num_replicas_in_sync)),
100 distribution=strategy_combinations.all_strategies,
103 def testStatefulExperimentalRunAlwaysExecute(self, distribution): argument
104 with distribution.scope():
114 distribution.run(assign_add)
122 distribution=strategy_combinations.strategies_minus_tpu,
[all …]
|
| D | one_device_strategy_test.py | 31 distribution=[
40 def testMinimizeLoss(self, distribution): argument
42 self._test_minimize_loss_eager(distribution)
44 self._test_minimize_loss_graph(distribution)
46 def testReplicaId(self, distribution): argument
47 self._test_replica_id(distribution)
49 def testCallAndMergeExceptions(self, distribution): argument
50 self._test_call_and_merge_exceptions(distribution)
52 def testReplicateDataset(self, distribution): argument
62 self._test_input_fn_iterable(distribution, input_fn, expected_values)
[all …]
|
| D | mirrored_variable_test.py | 63 distribution=[
103 def testVariableInFuncGraph(self, distribution): argument
110 with func_graph.FuncGraph("fg").as_default(), distribution.scope():
112 v2 = distribution.extended.call_for_each_replica(model_fn)
114 self._test_mv_properties(v1, "foo:0", distribution)
115 self._test_mv_properties(v2, "bar:0", distribution)
117 def testVariableWithTensorInitialValueInFunction(self, distribution): argument
132 return distribution.experimental_local_results(
133 distribution.extended.call_for_each_replica(model_fn))
137 def testSingleVariable(self, distribution): argument
[all …]
|
| D | vars_test.py | 52 # distribution=[
59 distribution=[
70 distribution=[
85 def testAssign(self, distribution, experimental_run_tf_function): argument
94 return test_util.gather(distribution, distribution.run(update_fn))
112 with distribution.scope():
124 def testAssignOnWriteVar(self, distribution, experimental_run_tf_function): argument
126 with distribution.scope():
139 return test_util.gather(distribution, distribution.run(update_fn))
157 with distribution.scope():
[all …]
|
| D | values_test.py | 58 distribution=[
72 distribution=(strategy_combinations.all_strategies_minus_default +
76 def testMakeDistributedValueFromTensor(self, distribution): argument
85 distribution.experimental_distribute_values_from_function(value_fn))
87 ds_test_util.gather(distribution, distributed_values),
88 constant_op.constant(1., shape=(distribution.num_replicas_in_sync)))
92 distribution=(strategy_combinations.all_strategies_minus_default +
96 def testMakeDistributedValueSingleNumpyArrayConstant(self, distribution): argument
105 distribution.experimental_distribute_values_from_function(value_fn))
107 ds_test_util.gather(distribution, distributed_values).numpy(),
[all …]
|
| D | input_lib_test.py | 287 distribution=[
290 def testMultiDeviceIterInitialize(self, distribution): argument
300 dataset_fn(distribute_lib.InputContext()), input_workers, distribution)
316 distribution=[
321 def testOneDeviceCPU(self, input_type, api_type, iteration_type, distribution, argument
330 distribution.extended.experimental_enable_get_next_as_optional = (
334 expected_values, distribution)
342 distribution=[strategy_combinations.multi_worker_mirrored_2x1_cpu],
345 distribution, enable_get_next_as_optional): argument
353 distribution.extended.experimental_enable_get_next_as_optional = (
[all …]
|
| D | metrics_v1_test.py | 73 distribution=[
84 distribution=[
95 def _test_metric(self, distribution, dataset_fn, metric_fn, expected_fn): argument
96 with ops.Graph().as_default(), distribution.scope():
97 iterator = distribution.make_input_fn_iterator(lambda _: dataset_fn())
98 if strategy_test_lib.is_tpu_strategy(distribution):
100 value, update = distribution.extended.call_for_each_replica(
103 return distribution.group(update)
105 ctx = distribution.extended.experimental_run_steps_on_iterator(
106 step_fn, iterator, iterations=distribution.extended.steps_per_run)
[all …]
|
| D | moving_averages_test.py | 44 distribution=all_distributions, mode=["graph"])
47 distribution=all_distributions, mode=["eager"], use_function=[True, False])
53 def testReplicaModeWithoutZeroDebias(self, distribution): argument
65 with distribution.scope():
66 var, assign = distribution.extended.call_for_each_replica(replica_fn)
69 self.evaluate(distribution.experimental_local_results(assign))
80 def testReplicaMode(self, distribution): argument
91 with distribution.scope():
92 var, assign_op = distribution.extended.call_for_each_replica(replica_fn)
95 self.evaluate(distribution.experimental_local_results(assign_op))
[all …]
|
| D | input_lib_type_spec_test.py | 51 distribution=[
56 def testTypeSpec(self, input_type, distribution, argument
63 distribution.extended.experimental_enable_get_next_as_optional = (
66 dist_dataset = distribution.experimental_distribute_dataset(dataset)
67 with distribution.scope():
83 distribution=[
89 distribution, enable_get_next_as_optional): argument
96 distribution.extended.experimental_enable_get_next_as_optional = (
99 dist_dataset = distribution.experimental_distribute_dataset(dataset)
100 with distribution.scope():
[all …]
|
| D | distributed_variable_test.py | 63 distribution=[
74 distribution=[
100 def testExtendsVariable(self, distribution, synchronization, aggregation): argument
101 with distribution.scope():
106 def testCheckpointing(self, distribution, synchronization, aggregation, mode): argument
108 if (isinstance(distribution,
113 with distribution.scope():
139 def testTraceback(self, distribution, synchronization, aggregation): argument
142 with distribution.scope():
158 def testSelectReplica(self, distribution, synchronization, aggregation): argument
[all …]
|
| D | tf_function_test.py | 15 """Tests for tf.function + distribution strategies."""
48 distribution=strategy_combinations.all_strategies,
53 self, distribution, run_functions_eagerly): argument
57 worker = distribution.extended.worker_devices[0]
62 with distribution.scope():
74 distribution=strategy_combinations.all_strategies,
79 self, distribution, run_functions_eagerly): argument
83 worker = distribution.extended.worker_devices[0]
88 with distribution.scope():
105 distribution=strategy_combinations.all_strategies,
[all …]
|
| /external/rust/crates/criterion/src/stats/ |
| D | tuple.rs | 3 use crate::stats::Distribution;
14 /// A tuple of distributions: `(Distribution<A>, Distribution<B>, ..)`
42 type Distributions = (Distribution<A>,);
46 impl<A> TupledDistributions for (Distribution<A>,) implementation
70 fn complete(self) -> (Distribution<A>,) { in complete()
71 (Distribution(self.0.into_boxed_slice()),) in complete()
80 type Distributions = (Distribution<A>, Distribution<B>);
84 impl<A, B> TupledDistributions for (Distribution<A>, Distribution<B>) implementation
112 fn complete(self) -> (Distribution<A>, Distribution<B>) { in complete()
114 Distribution(self.0.into_boxed_slice()), in complete()
[all …]
|
| /external/tensorflow/tensorflow/python/ops/ |
| D | nn_loss_scaling_utilities_test.py | 69 distribution=[
73 def testComputeAverageLossDefaultGlobalBatchSize(self, distribution): argument
80 with distribution.scope():
81 per_replica_losses = distribution.run(
83 loss = distribution.reduce("SUM", per_replica_losses, axis=None)
88 distribution=[
92 def testComputeAverageLossSampleWeights(self, distribution): argument
93 with distribution.scope():
95 per_replica_losses = distribution.run(
99 loss = distribution.reduce("SUM", per_replica_losses, axis=None)
[all …]
|
| /external/tensorflow/tensorflow/python/ops/distributions/ |
| D | transformed_distribution.py | 15 """A Transformed Distribution class."""
26 from tensorflow.python.ops.distributions import distribution as distribution_lib
118 class TransformedDistribution(distribution_lib.Distribution):
119 """A Transformed Distribution.
121 A `TransformedDistribution` models `p(y)` given a base distribution `p(x)`,
124 distribution is typically an instance of the `Distribution` class.
133 `Distribution` associated with a random variable (rv) `X`.
135 Write `cdf(Y=y)` for an absolutely continuous cumulative distribution function
147 Programmatically: `bijector.forward(distribution.sample(...))`
152 Programmatically: `(distribution.log_prob(bijector.inverse(y))
[all …]
|
| /external/python/setuptools/pkg_resources/ |
| D | api_tests.txt | 7 A "Distribution" is a collection of files that represent a "Release" of a
12 >>> from pkg_resources import Distribution
13 >>> Distribution(project_name="Foo", version="1.2")
19 >>> dist = Distribution(
56 >>> Distribution(version='1.0') == Distribution(version='1.0')
58 >>> Distribution(version='1.0') == Distribution(version='1.1')
60 >>> Distribution(version='1.0') < Distribution(version='1.1')
66 >>> Distribution(project_name="Foo",version="1.0") == \
67 ... Distribution(project_name="Foo",version="1.0")
70 >>> Distribution(project_name="Foo",version="1.0") == \
[all …]
|
| /external/tensorflow/tensorflow/core/kernels/ |
| D | random_op_gpu.h | 31 template <class Distribution, bool VariableSamplesPerOutput>
34 template <class Distribution>
35 struct FillPhiloxRandomKernel<Distribution, false> {
36 typedef typename Distribution::ResultElementType T;
39 Distribution dist);
42 template <class Distribution>
43 struct FillPhiloxRandomKernel<Distribution, true> {
44 typedef typename Distribution::ResultElementType T;
47 int64 size, Distribution dist);
139 // distribution. Each output takes a fixed number of samples.
[all …]
|
| D | random_op_cpu.h | 60 template <typename Device, class Distribution>
62 typedef typename Distribution::ResultElementType T;
65 int64_t size, Distribution dist) { in operator()
71 "not support this device or random distribution yet.")); in operator()
76 template <class Distribution, bool VariableSamplesPerOutput>
79 // Specialization for distribution that takes a fixed number of samples for
81 template <class Distribution>
82 struct FillPhiloxRandomTask<Distribution, false> {
83 typedef typename Distribution::ResultElementType T;
85 int64_t start_group, int64_t limit_group, Distribution dist) {
[all …]
|
| /external/tensorflow/tensorflow/tools/api/golden/v1/ |
| D | tensorflow.distributions.pbtxt | 5 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
9 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
13 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
17 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
21 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
24 name: "Distribution"
25 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
29 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
33 mtype: "<class \'tensorflow.python.ops.distributions.distribution.ReparameterizationType\'>"
37 mtype: "<class \'tensorflow.python.ops.distributions.distribution._DistributionMeta\'>"
[all …]
|
| /external/python/setuptools/docs/ |
| D | pkg_resources.rst | 33 Eggs are a distribution format for Python modules, similar in concept to
35 However, unlike a pure distribution format, eggs can also be installed and
42 a distribution to co-exist in the same Python installation, with individual
60 distribution
63 importable distribution
67 pluggable distribution
68 An importable distribution whose filename unambiguously identifies its
82 necessarily active. More than one distribution (i.e. release version) for
87 ``sys.path``. At most one distribution (release version) of a given
99 default version of a distribution that is available to software that
[all …]
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/distribution/ |
| D | PoissonDistributionImpl.java | 17 package org.apache.commons.math.distribution;
51 /** Distribution used to compute normal approximation. */
55 * Holds the Poisson mean for the distribution.
73 * Create a new Poisson distribution with the given the mean. The mean value
84 * Create a new Poisson distribution with the given mean, convergence criterion
99 * Create a new Poisson distribution with the given mean and convergence criterion.
111 * Create a new Poisson distribution with the given mean and maximum number of iterations.
124 * Create a new Poisson distribution with the given the mean. The mean value
128 * @param z a normal distribution used to compute normal approximations.
141 * Get the Poisson mean for the distribution.
[all …]
|
| /external/apache-commons-math/src/main/java/org/apache/commons/math/random/ |
| D | RandomDataImpl.java | 28 import org.apache.commons.math.distribution.BetaDistributionImpl;
29 import org.apache.commons.math.distribution.BinomialDistributionImpl;
30 import org.apache.commons.math.distribution.CauchyDistributionImpl;
31 import org.apache.commons.math.distribution.ChiSquaredDistributionImpl;
32 import org.apache.commons.math.distribution.ContinuousDistribution;
33 import org.apache.commons.math.distribution.FDistributionImpl;
34 import org.apache.commons.math.distribution.GammaDistributionImpl;
35 import org.apache.commons.math.distribution.HypergeometricDistributionImpl;
36 import org.apache.commons.math.distribution.IntegerDistribution;
37 import org.apache.commons.math.distribution.PascalDistributionImpl;
[all …]
|
| D | EmpiricalDistribution.java | 30 * empirical probability distribution</a> -- a probability distribution derived
32 * of the population distribution that the data come from.<p>
34 * <i>distribution digests</i>, that describe empirical distributions and
36 * <li>loading the distribution from a file of observed data values</li>
41 * <li>generating random values from the distribution</li>
46 * generated will follow the distribution of the values in the file.</p>
53 * Computes the empirical distribution from the provided
61 * Computes the empirical distribution from the input file.
69 * Computes the empirical distribution using data read from a URL.
77 * Generates a random value from this distribution.
[all …]
|
| /external/rust/crates/criterion/src/ |
| D | estimate.rs | 3 use crate::stats::Distribution;
50 let to_estimate = |point_estimate, distribution: &Distribution<f64>| { in build_estimates()
51 let (lb, ub) = distribution.confidence_interval(cl); in build_estimates()
60 standard_error: distribution.std_dev(None), in build_estimates()
78 let to_estimate = |point_estimate, distribution: &Distribution<f64>| { in build_change_estimates()
79 let (lb, ub) = distribution.confidence_interval(cl); in build_change_estimates()
88 standard_error: distribution.std_dev(None), in build_change_estimates()
130 pub mean: Distribution<f64>,
131 pub median: Distribution<f64>,
132 pub median_abs_dev: Distribution<f64>,
[all …]
|
| /external/rust/crates/rand/src/distributions/ |
| D | mod.rs | 12 //! This module is the home of the [`Distribution`] trait and several of its
17 //! Abstractly, a [probability distribution] describes the probability of
20 //! More concretely, an implementation of `Distribution<T>` for type `X` is an
22 //! according to the distribution `X` represents, using an external source of
25 //! A type `X` may implement `Distribution<T>` for multiple types `T`.
26 //! Any type implementing [`Distribution`] is stateless (i.e. immutable),
31 //! # The `Standard` distribution
33 //! The [`Standard`] distribution is important to mention. This is the
34 //! distribution used by [`Rng::gen`] and represents the "default" way to
39 //! Implementing `Distribution<T>` for [`Standard`] for user types `T` makes it
[all …]
|