experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Update generated Python Op docs.

Browse Source 
Change: 141516835
This commit is contained in:
A. Unique TensorFlower 2016-12-08 20:22:11 -08:00 committed by TensorFlower Gardener
parent 2709e16b98
commit 4a29f9f6a0
30 changed files with 1139 additions and 1072 deletions

26
tensorflow/g3doc/api_docs/python/contrib.learn.md
View File

@ -729,17 +729,17 @@ Trains a model given training data `x` predictions and `y` labels.
##### Args:
* <b>`x`</b>: Matrix of shape [n_samples, n_features...]. Can be iterator that
returns arrays of features. The training input samples for fitting the
model. If set, `input_fn` must be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs]. Can be
iterator that returns array of labels. The training label values
(class labels in classification, real numbers in regression). If set,
`input_fn` must be `None`. Note: For classification, label values must
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] or the dictionary of Matrices.
Can be iterator that returns arrays of features or dictionary of arrays of features.
The training input samples for fitting the model. If set, `input_fn` must be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs] or the dictionary of same.
Can be iterator that returns array of labels or dictionary of array of labels.
The training label values (class labels in classification, real numbers in regression).
If set, `input_fn` must be `None`. Note: For classification, label values must
be integers representing the class index (i.e. values from 0 to
n_classes-1).
* <b>`input_fn`</b>: Input function returning a tuple of:
features - Dictionary of string feature name to `Tensor` or `Tensor`.
features - `Tensor` or dictionary of string feature name to `Tensor`.
labels - `Tensor` or dictionary of `Tensor` with labels.
If input_fn is set, `x`, `y`, and `batch_size` must be `None`.
* <b>`steps`</b>: Number of steps for which to train model. If `None`, train forever.
@ -791,12 +791,14 @@ for which this evaluation was performed.
##### Args:
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] containing the input samples
for fitting the model. Can be iterator that returns arrays of features.
If set, `input_fn` must be `None`.
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] or dictionary of many matrices
containing the input samples for fitting the model. Can be iterator that returns
arrays of features or dictionary of array of features. If set, `input_fn` must
be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs] containing the
label values (class labels in classification, real numbers in
regression). Can be iterator that returns array of labels. If set,
regression) or dictionary of multiple vectors/matrices. Can be iterator
that returns array of targets or dictionary of array of targets. If set,
`input_fn` must be `None`. Note: For classification, label values must
be integers representing the class index (i.e. values from 0 to
n_classes-1).

2
tensorflow/g3doc/api_docs/python/control_flow_ops.md
View File

@ -186,7 +186,7 @@ it has occasionally surprised some users who expected a lazier semantics.
y = tf.constant(5)
def f1(): return tf.mul(x, 17)
def f2(): return tf.add(y, 23)
r = cond(tf.less(x, y), f1, f2)
r = tf.cond(tf.less(x, y), f1, f2)
# r is set to f1().
# Operations in f2 (e.g., tf.add) are not executed.
```

11
tensorflow/g3doc/api_docs/python/framework.md
View File

@ -2835,17 +2835,6 @@ The following standard keys are defined:
* `WEIGHTS`: weights inside neural network layers
* `BIASES`: biases inside neural network layers
* `ACTIVATIONS`: activations of neural network layers
- - -
#### `tf.GraphKeys.VARIABLES` {#GraphKeys.VARIABLES}
DEPRECATED FUNCTION
THIS FUNCTION IS DEPRECATED. It will be removed after 2017-03-02.
Instructions for updating:
VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead
## Defining new operations

2
tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.cond.md
View File

@ -47,7 +47,7 @@ it has occasionally surprised some users who expected a lazier semantics.
y = tf.constant(5)
def f1(): return tf.mul(x, 17)
def f2(): return tf.add(y, 23)
r = cond(tf.less(x, y), f1, f2)
r = tf.cond(tf.less(x, y), f1, f2)
# r is set to f1().
# Operations in f2 (e.g., tf.add) are not executed.
```

244
tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.summary.TaggedRunMetadata.md
View File

@ -1,185 +1,4 @@
- - -
#### `tf.summary.TaggedRunMetadata.ByteSize()` {#TaggedRunMetadata.ByteSize}
- - -
#### `tf.summary.TaggedRunMetadata.Clear()` {#TaggedRunMetadata.Clear}
- - -
#### `tf.summary.TaggedRunMetadata.ClearExtension(extension_handle)` {#TaggedRunMetadata.ClearExtension}
- - -
#### `tf.summary.TaggedRunMetadata.ClearField(field_name)` {#TaggedRunMetadata.ClearField}
- - -
#### `tf.summary.TaggedRunMetadata.CopyFrom(other_msg)` {#TaggedRunMetadata.CopyFrom}
Copies the content of the specified message into the current message.
The method clears the current message and then merges the specified
message using MergeFrom.
##### Args:
* <b>`other_msg`</b>: Message to copy into the current one.
- - -
#### `tf.summary.TaggedRunMetadata.DiscardUnknownFields()` {#TaggedRunMetadata.DiscardUnknownFields}
- - -
#### `tf.summary.TaggedRunMetadata.FindInitializationErrors()` {#TaggedRunMetadata.FindInitializationErrors}
Finds required fields which are not initialized.
##### Returns:
A list of strings. Each string is a path to an uninitialized field from
the top-level message, e.g. "foo.bar[5].baz".
- - -
#### `tf.summary.TaggedRunMetadata.FromString(s)` {#TaggedRunMetadata.FromString}
- - -
#### `tf.summary.TaggedRunMetadata.HasExtension(extension_handle)` {#TaggedRunMetadata.HasExtension}
- - -
#### `tf.summary.TaggedRunMetadata.HasField(field_name)` {#TaggedRunMetadata.HasField}
- - -
#### `tf.summary.TaggedRunMetadata.IsInitialized(errors=None)` {#TaggedRunMetadata.IsInitialized}
Checks if all required fields of a message are set.
##### Args:
* <b>`errors`</b>: A list which, if provided, will be populated with the field
paths of all missing required fields.
##### Returns:
True iff the specified message has all required fields set.
- - -
#### `tf.summary.TaggedRunMetadata.ListFields()` {#TaggedRunMetadata.ListFields}
- - -
#### `tf.summary.TaggedRunMetadata.MergeFrom(msg)` {#TaggedRunMetadata.MergeFrom}
- - -
#### `tf.summary.TaggedRunMetadata.MergeFromString(serialized)` {#TaggedRunMetadata.MergeFromString}
- - -
#### `tf.summary.TaggedRunMetadata.ParseFromString(serialized)` {#TaggedRunMetadata.ParseFromString}
Parse serialized protocol buffer data into this message.
Like MergeFromString(), except we clear the object first and
do not return the value that MergeFromString returns.
- - -
#### `tf.summary.TaggedRunMetadata.RegisterExtension(extension_handle)` {#TaggedRunMetadata.RegisterExtension}
- - -
#### `tf.summary.TaggedRunMetadata.SerializePartialToString()` {#TaggedRunMetadata.SerializePartialToString}
- - -
#### `tf.summary.TaggedRunMetadata.SerializeToString()` {#TaggedRunMetadata.SerializeToString}
- - -
#### `tf.summary.TaggedRunMetadata.SetInParent()` {#TaggedRunMetadata.SetInParent}
Sets the _cached_byte_size_dirty bit to true,
and propagates this to our listener iff this was a state change.
- - -
#### `tf.summary.TaggedRunMetadata.WhichOneof(oneof_name)` {#TaggedRunMetadata.WhichOneof}
Returns the name of the currently set field inside a oneof, or None.
- - -
#### `tf.summary.TaggedRunMetadata.__deepcopy__(memo=None)` {#TaggedRunMetadata.__deepcopy__}
- - -
#### `tf.summary.TaggedRunMetadata.__eq__(other)` {#TaggedRunMetadata.__eq__}
- - -
#### `tf.summary.TaggedRunMetadata.__getstate__()` {#TaggedRunMetadata.__getstate__}
@ -187,66 +6,3 @@ Returns the name of the currently set field inside a oneof, or None.
Support the pickle protocol.
- - -
#### `tf.summary.TaggedRunMetadata.__hash__()` {#TaggedRunMetadata.__hash__}
- - -
#### `tf.summary.TaggedRunMetadata.__init__(**kwargs)` {#TaggedRunMetadata.__init__}
- - -
#### `tf.summary.TaggedRunMetadata.__ne__(other_msg)` {#TaggedRunMetadata.__ne__}
- - -
#### `tf.summary.TaggedRunMetadata.__repr__()` {#TaggedRunMetadata.__repr__}
- - -
#### `tf.summary.TaggedRunMetadata.__setstate__(state)` {#TaggedRunMetadata.__setstate__}
Support the pickle protocol.
- - -
#### `tf.summary.TaggedRunMetadata.__str__()` {#TaggedRunMetadata.__str__}
- - -
#### `tf.summary.TaggedRunMetadata.__unicode__()` {#TaggedRunMetadata.__unicode__}
- - -
#### `tf.summary.TaggedRunMetadata.run_metadata` {#TaggedRunMetadata.run_metadata}
Magic attribute generated for "run_metadata" proto field.
- - -
#### `tf.summary.TaggedRunMetadata.tag` {#TaggedRunMetadata.tag}
Magic attribute generated for "tag" proto field.

2
tensorflow/g3doc/api_docs/python/functions_and_classes/shard0/tf.train.ExponentialMovingAverage.md
View File

@ -121,7 +121,7 @@ debiased (see docstring in `assign_moving_average` for more details).
shadow variables are created with `trainable=False` and added to the
`GraphKeys.ALL_VARIABLES` collection. They will be returned by calls to
`tf.all_variables()`.
`tf.global_variables()`.
Returns an op that updates all shadow variables as described above.

3
tensorflow/g3doc/api_docs/python/functions_and_classes/shard1/tf.nn.embedding_lookup_sparse.md
View File

@ -15,7 +15,8 @@ is the sum of the size of params along dimension 0.
* <b>`params`</b>: A single tensor representing the complete embedding tensor,
or a list of P tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0.
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
* <b>`sp_ids`</b>: N x M SparseTensor of int64 ids (typically from FeatureValueToId),
where N is typically batch size and M is arbitrary.
* <b>`sp_weights`</b>: either a SparseTensor of float / double weights, or None to

9
tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.embedding_lookup.md
View File

@ -30,10 +30,11 @@ tensor. The returned tensor has shape `shape(ids) + shape(params)[1:]`.
##### Args:
* <b>`params`</b>: A list of tensors with the same type and which can be concatenated
along dimension 0. Alternatively, a `PartitionedVariable`, created by
partitioning along dimension 0. Each element must be appropriately sized
for the given `partition_strategy`.
* <b>`params`</b>: A single tensor representing the complete embedding tensor,
or a list of P tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
* <b>`ids`</b>: A `Tensor` with type `int32` or `int64` containing the ids to be looked
up in `params`.
* <b>`partition_strategy`</b>: A string specifying the partitioning strategy, relevant

2
tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.scatter_update.md
View File

@ -17,7 +17,7 @@ This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
If values in `ref` is to be updated more than once, because there are
duplicate entires in `indices`, the order at which the updates happen
duplicate entries in `indices`, the order at which the updates happen
for each value is undefined.
Requires `updates.shape = indices.shape + ref.shape[1:]`.

237
tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.summary.SummaryDescription.md
View File

@ -1,185 +1,4 @@
- - -
#### `tf.summary.SummaryDescription.ByteSize()` {#SummaryDescription.ByteSize}
- - -
#### `tf.summary.SummaryDescription.Clear()` {#SummaryDescription.Clear}
- - -
#### `tf.summary.SummaryDescription.ClearExtension(extension_handle)` {#SummaryDescription.ClearExtension}
- - -
#### `tf.summary.SummaryDescription.ClearField(field_name)` {#SummaryDescription.ClearField}
- - -
#### `tf.summary.SummaryDescription.CopyFrom(other_msg)` {#SummaryDescription.CopyFrom}
Copies the content of the specified message into the current message.
The method clears the current message and then merges the specified
message using MergeFrom.
##### Args:
* <b>`other_msg`</b>: Message to copy into the current one.
- - -
#### `tf.summary.SummaryDescription.DiscardUnknownFields()` {#SummaryDescription.DiscardUnknownFields}
- - -
#### `tf.summary.SummaryDescription.FindInitializationErrors()` {#SummaryDescription.FindInitializationErrors}
Finds required fields which are not initialized.
##### Returns:
A list of strings. Each string is a path to an uninitialized field from
the top-level message, e.g. "foo.bar[5].baz".
- - -
#### `tf.summary.SummaryDescription.FromString(s)` {#SummaryDescription.FromString}
- - -
#### `tf.summary.SummaryDescription.HasExtension(extension_handle)` {#SummaryDescription.HasExtension}
- - -
#### `tf.summary.SummaryDescription.HasField(field_name)` {#SummaryDescription.HasField}
- - -
#### `tf.summary.SummaryDescription.IsInitialized(errors=None)` {#SummaryDescription.IsInitialized}
Checks if all required fields of a message are set.
##### Args:
* <b>`errors`</b>: A list which, if provided, will be populated with the field
paths of all missing required fields.
##### Returns:
True iff the specified message has all required fields set.
- - -
#### `tf.summary.SummaryDescription.ListFields()` {#SummaryDescription.ListFields}
- - -
#### `tf.summary.SummaryDescription.MergeFrom(msg)` {#SummaryDescription.MergeFrom}
- - -
#### `tf.summary.SummaryDescription.MergeFromString(serialized)` {#SummaryDescription.MergeFromString}
- - -
#### `tf.summary.SummaryDescription.ParseFromString(serialized)` {#SummaryDescription.ParseFromString}
Parse serialized protocol buffer data into this message.
Like MergeFromString(), except we clear the object first and
do not return the value that MergeFromString returns.
- - -
#### `tf.summary.SummaryDescription.RegisterExtension(extension_handle)` {#SummaryDescription.RegisterExtension}
- - -
#### `tf.summary.SummaryDescription.SerializePartialToString()` {#SummaryDescription.SerializePartialToString}
- - -
#### `tf.summary.SummaryDescription.SerializeToString()` {#SummaryDescription.SerializeToString}
- - -
#### `tf.summary.SummaryDescription.SetInParent()` {#SummaryDescription.SetInParent}
Sets the _cached_byte_size_dirty bit to true,
and propagates this to our listener iff this was a state change.
- - -
#### `tf.summary.SummaryDescription.WhichOneof(oneof_name)` {#SummaryDescription.WhichOneof}
Returns the name of the currently set field inside a oneof, or None.
- - -
#### `tf.summary.SummaryDescription.__deepcopy__(memo=None)` {#SummaryDescription.__deepcopy__}
- - -
#### `tf.summary.SummaryDescription.__eq__(other)` {#SummaryDescription.__eq__}
- - -
#### `tf.summary.SummaryDescription.__getstate__()` {#SummaryDescription.__getstate__}
@ -187,59 +6,3 @@ Returns the name of the currently set field inside a oneof, or None.
Support the pickle protocol.
- - -
#### `tf.summary.SummaryDescription.__hash__()` {#SummaryDescription.__hash__}
- - -
#### `tf.summary.SummaryDescription.__init__(**kwargs)` {#SummaryDescription.__init__}
- - -
#### `tf.summary.SummaryDescription.__ne__(other_msg)` {#SummaryDescription.__ne__}
- - -
#### `tf.summary.SummaryDescription.__repr__()` {#SummaryDescription.__repr__}
- - -
#### `tf.summary.SummaryDescription.__setstate__(state)` {#SummaryDescription.__setstate__}
Support the pickle protocol.
- - -
#### `tf.summary.SummaryDescription.__str__()` {#SummaryDescription.__str__}
- - -
#### `tf.summary.SummaryDescription.__unicode__()` {#SummaryDescription.__unicode__}
- - -
#### `tf.summary.SummaryDescription.type_hint` {#SummaryDescription.type_hint}
Magic attribute generated for "type_hint" proto field.

521
tensorflow/g3doc/api_docs/python/functions_and_classes/shard2/tf.test.TestCase.md
View File

@ -173,6 +173,125 @@ Checks that for all elements of farray1 and farray2
* <b>`err`</b>: a float value.
- - -
#### `tf.test.TestCase.assertBetween(value, minv, maxv, msg=None)` {#TestCase.assertBetween}
Asserts that value is between minv and maxv (inclusive).
- - -
#### `tf.test.TestCase.assertCommandFails(command, regexes, env=None, close_fds=True, msg=None)` {#TestCase.assertCommandFails}
Asserts a shell command fails and the error matches a regex in a list.
##### Args:
* <b>`command`</b>: List or string representing the command to run.
* <b>`regexes`</b>: the list of regular expression strings.
* <b>`env`</b>: Dictionary of environment variable settings.
* <b>`close_fds`</b>: Whether or not to close all open fd's in the child after
forking.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertCommandSucceeds(command, regexes=('',), env=None, close_fds=True, msg=None)` {#TestCase.assertCommandSucceeds}
Asserts that a shell command succeeds (i.e. exits with code 0).
##### Args:
* <b>`command`</b>: List or string representing the command to run.
* <b>`regexes`</b>: List of regular expression byte strings that match success.
* <b>`env`</b>: Dictionary of environment variable settings.
* <b>`close_fds`</b>: Whether or not to close all open fd's in the child after
forking.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsExactSubsequence(container, subsequence, msg=None)` {#TestCase.assertContainsExactSubsequence}
Assert that "container" contains "subsequence" as an exact subsequence.
Asserts that "container" contains all the elements of "subsequence", in
order, and without other elements interspersed. For example, [1, 2, 3] is an
exact subsequence of [0, 0, 1, 2, 3, 0] but not of [0, 0, 1, 2, 0, 3, 0].
##### Args:
* <b>`container`</b>: the list we're testing for subsequence inclusion.
* <b>`subsequence`</b>: the list we hope will be an exact subsequence of container.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsInOrder(strings, target, msg=None)` {#TestCase.assertContainsInOrder}
Asserts that the strings provided are found in the target in order.
This may be useful for checking HTML output.
##### Args:
* <b>`strings`</b>: A list of strings, such as [ 'fox', 'dog' ]
* <b>`target`</b>: A target string in which to look for the strings, such as
'The quick brown fox jumped over the lazy dog'.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsSubsequence(container, subsequence, msg=None)` {#TestCase.assertContainsSubsequence}
Assert that "container" contains "subsequence" as a subsequence.
Asserts that "container" contains all the elements of "subsequence", in
order, but possibly with other elements interspersed. For example, [1, 2, 3]
is a subsequence of [0, 0, 1, 2, 0, 3, 0] but not of [0, 0, 1, 3, 0, 2, 0].
##### Args:
* <b>`container`</b>: the list we're testing for subsequence inclusion.
* <b>`subsequence`</b>: the list we hope will be a subsequence of container.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsSubset(expected_subset, actual_set, msg=None)` {#TestCase.assertContainsSubset}
Checks whether actual iterable is a superset of expected iterable.
- - -
#### `tf.test.TestCase.assertCountEqual(*args, **kwargs)` {#TestCase.assertCountEqual}
An unordered sequence specific comparison.
Equivalent to assertItemsEqual(). This method is a compatibility layer
for Python 3k, since 2to3 does not convert assertItemsEqual() calls into
assertCountEqual() calls.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertDeviceEqual(device1, device2)` {#TestCase.assertDeviceEqual}
@ -195,9 +314,48 @@ Checks whether actual is a superset of expected.
- - -
#### `tf.test.TestCase.assertDictEqual(d1, d2, msg=None)` {#TestCase.assertDictEqual}
#### `tf.test.TestCase.assertDictEqual(a, b, msg=None)` {#TestCase.assertDictEqual}
Raises AssertionError if a and b are not equal dictionaries.
##### Args:
* <b>`a`</b>: A dict, the expected value.
* <b>`b`</b>: A dict, the actual value.
* <b>`msg`</b>: An optional str, the associated message.
##### Raises:
* <b>`AssertionError`</b>: if the dictionaries are not equal.
- - -
#### `tf.test.TestCase.assertEmpty(container, msg=None)` {#TestCase.assertEmpty}
Assert that an object has zero length.
##### Args:
* <b>`container`</b>: Anything that implements the collections.Sized interface.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertEndsWith(actual, expected_end, msg=None)` {#TestCase.assertEndsWith}
Assert that actual.endswith(expected_end) is True.
##### Args:
* <b>`actual`</b>: str
* <b>`expected_end`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
@ -282,10 +440,11 @@ Included for symmetry with assertIsNone.
- - -
#### `tf.test.TestCase.assertItemsEqual(expected_seq, actual_seq, msg=None)` {#TestCase.assertItemsEqual}
#### `tf.test.TestCase.assertItemsEqual(*args, **kwargs)` {#TestCase.assertItemsEqual}
An unordered sequence specific comparison. It asserts that
actual_seq and expected_seq have the same element counts.
An unordered sequence specific comparison.
It asserts that actual_seq and expected_seq have the same element counts.
Equivalent to::
self.assertEqual(Counter(iter(actual_seq)),
@ -298,6 +457,30 @@ Asserts that each element has the same count in both sequences.
- [0, 1, 1] and [1, 0, 1] compare equal.
- [0, 0, 1] and [0, 1] compare unequal.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertJsonEqual(first, second, msg=None)` {#TestCase.assertJsonEqual}
Asserts that the JSON objects defined in two strings are equal.
A summary of the differences will be included in the failure message
using assertSameStructure.
##### Args:
* <b>`first`</b>: A string contining JSON to decode and compare to second.
* <b>`second`</b>: A string contining JSON to decode and compare to first.
* <b>`msg`</b>: Additional text to include in the failure message.
- - -
@ -367,6 +550,13 @@ if not.
* <b>`msg`</b>: An optional string message to append to the failure message.
- - -
#### `tf.test.TestCase.assertNoCommonElements(expected_seq, actual_seq, msg=None)` {#TestCase.assertNoCommonElements}
Checks whether actual iterable and expected iterable are disjoint.
- - -
#### `tf.test.TestCase.assertNotAlmostEqual(first, second, places=None, msg=None, delta=None)` {#TestCase.assertNotAlmostEqual}
@ -397,6 +587,33 @@ as significant digits (measured from the most signficant digit).
Objects that are equal automatically fail.
- - -
#### `tf.test.TestCase.assertNotEmpty(container, msg=None)` {#TestCase.assertNotEmpty}
Assert that an object has non-zero length.
##### Args:
* <b>`container`</b>: Anything that implements the collections.Sized interface.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertNotEndsWith(actual, unexpected_end, msg=None)` {#TestCase.assertNotEndsWith}
Assert that actual.endswith(unexpected_end) is False.
##### Args:
* <b>`actual`</b>: str
* <b>`unexpected_end`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertNotEqual(first, second, msg=None)` {#TestCase.assertNotEqual}
@ -434,6 +651,20 @@ Included for symmetry with assertIsInstance.
Fail the test if the text matches the regular expression.
- - -
#### `tf.test.TestCase.assertNotStartsWith(actual, unexpected_start, msg=None)` {#TestCase.assertNotStartsWith}
Assert that actual.startswith(unexpected_start) is False.
##### Args:
* <b>`actual`</b>: str
* <b>`unexpected_start`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertProtoEquals(expected_message_maybe_ascii, message)` {#TestCase.assertProtoEquals}
@ -508,6 +739,38 @@ Asserts that the message in a raised exception matches a regexp.
* <b>`kwargs`</b>: Extra kwargs.
- - -
#### `tf.test.TestCase.assertRaisesWithLiteralMatch(expected_exception, expected_exception_message, callable_obj=None, *args, **kwargs)` {#TestCase.assertRaisesWithLiteralMatch}
Asserts that the message in a raised exception equals the given string.
Unlike assertRaisesRegexp, this method takes a literal string, not
a regular expression.
with self.assertRaisesWithLiteralMatch(ExType, 'message'):
DoSomething()
##### Args:
* <b>`expected_exception`</b>: Exception class expected to be raised.
* <b>`expected_exception_message`</b>: String message expected in the raised
exception. For a raise exception e, expected_exception_message must
equal str(e).
* <b>`callable_obj`</b>: Function to be called, or None to return a context.
* <b>`args`</b>: Extra args.
* <b>`kwargs`</b>: Extra kwargs.
##### Returns:
A context manager if callable_obj is None. Otherwise, None.
##### Raises:
self.failureException if callable_obj does not raise a macthing exception.
- - -
#### `tf.test.TestCase.assertRaisesWithPredicateMatch(exception_type, expected_err_re_or_predicate)` {#TestCase.assertRaisesWithPredicateMatch}
@ -532,6 +795,71 @@ predicate search.
exception.
- - -
#### `tf.test.TestCase.assertRaisesWithRegexpMatch(expected_exception, expected_regexp, callable_obj=None, *args, **kwargs)` {#TestCase.assertRaisesWithRegexpMatch}
Asserts that the message in a raised exception matches the given regexp.
This is just a wrapper around assertRaisesRegexp. Please use
assertRaisesRegexp instead of assertRaisesWithRegexpMatch.
##### Args:
* <b>`expected_exception`</b>: Exception class expected to be raised.
* <b>`expected_regexp`</b>: Regexp (re pattern object or string) expected to be
found in error message.
* <b>`callable_obj`</b>: Function to be called, or None to return a context.
* <b>`args`</b>: Extra args.
* <b>`kwargs`</b>: Extra keyword args.
##### Returns:
A context manager if callable_obj is None. Otherwise, None.
##### Raises:
self.failureException if callable_obj does not raise a macthing exception.
- - -
#### `tf.test.TestCase.assertRegexMatch(actual_str, regexes, message=None)` {#TestCase.assertRegexMatch}
Asserts that at least one regex in regexes matches str.
If possible you should use assertRegexpMatches, which is a simpler
version of this method. assertRegexpMatches takes a single regular
expression (a string or re compiled object) instead of a list.
Notes:
1. This function uses substring matching, i.e. the matching
succeeds if *any* substring of the error message matches *any*
regex in the list. This is more convenient for the user than
full-string matching.
2. If regexes is the empty list, the matching will always fail.
3. Use regexes=[''] for a regex that will always pass.
4. '.' matches any single character *except* the newline. To
match any character, use '(.|
)'.
5. '^' matches the beginning of each line, not just the beginning
of the string. Similarly, '$' matches the end of each line.
6. An exception will be thrown if regexes contains an invalid
regex.
Args:
actual_str: The string we try to match with the items in regexes.
regexes: The regular expressions we want to match against str.
See "Notes" above for detailed notes on how this is interpreted.
message: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertRegexpMatches(text, expected_regexp, msg=None)` {#TestCase.assertRegexpMatches}
@ -539,6 +867,79 @@ predicate search.
Fail the test unless the text matches the regular expression.
- - -
#### `tf.test.TestCase.assertSameElements(expected_seq, actual_seq, msg=None)` {#TestCase.assertSameElements}
Assert that two sequences have the same elements (in any order).
This method, unlike assertItemsEqual, doesn't care about any
duplicates in the expected and actual sequences.
>> assertSameElements([1, 1, 1, 0, 0, 0], [0, 1])
# Doesn't raise an AssertionError
If possible, you should use assertItemsEqual instead of
assertSameElements.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertSameStructure(a, b, aname='a', bname='b', msg=None)` {#TestCase.assertSameStructure}
Asserts that two values contain the same structural content.
The two arguments should be data trees consisting of trees of dicts and
lists. They will be deeply compared by walking into the contents of dicts
and lists; other items will be compared using the == operator.
If the two structures differ in content, the failure message will indicate
the location within the structures where the first difference is found.
This may be helpful when comparing large structures.
##### Args:
* <b>`a`</b>: The first structure to compare.
* <b>`b`</b>: The second structure to compare.
* <b>`aname`</b>: Variable name to use for the first structure in assertion messages.
* <b>`bname`</b>: Variable name to use for the second structure.
* <b>`msg`</b>: Additional text to include in the failure message.
- - -
#### `tf.test.TestCase.assertSequenceAlmostEqual(expected_seq, actual_seq, places=None, msg=None, delta=None)` {#TestCase.assertSequenceAlmostEqual}
An approximate equality assertion for ordered sequences.
Fail if the two sequences are unequal as determined by their value
differences rounded to the given number of decimal places (default 7) and
comparing to zero, or by comparing that the difference between each value
in the two sequences is more than the given delta.
Note that decimal places (from zero) are usually not the same as significant
digits (measured from the most signficant digit).
If the two sequences compare equal then they will automatically compare
almost equal.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`places`</b>: The number of decimal places to compare.
* <b>`msg`</b>: The message to be printed if the test fails.
* <b>`delta`</b>: The OK difference between compared values.
- - -
#### `tf.test.TestCase.assertSequenceEqual(seq1, seq2, msg=None, seq_type=None)` {#TestCase.assertSequenceEqual}
@ -559,6 +960,26 @@ which can be indexed, has a length, and has an equality operator.
differences.
- - -
#### `tf.test.TestCase.assertSequenceStartsWith(prefix, whole, msg=None)` {#TestCase.assertSequenceStartsWith}
An equality assertion for the beginning of ordered sequences.
If prefix is an empty sequence, it will raise an error unless whole is also
an empty sequence.
If prefix is not a sequence, it will raise an error if the first element of
whole does not match.
##### Args:
* <b>`prefix`</b>: A sequence expected at the beginning of the whole parameter.
* <b>`whole`</b>: The sequence in which to look for prefix.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertSetEqual(set1, set2, msg=None)` {#TestCase.assertSetEqual}
@ -610,6 +1031,51 @@ Assert that actual.startswith(expected_start) is True.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertTotallyOrdered(*groups, **kwargs)` {#TestCase.assertTotallyOrdered}
Asserts that total ordering has been implemented correctly.
For example, say you have a class A that compares only on its attribute x.
Comparators other than __lt__ are omitted for brevity.
class A(object):
def __init__(self, x, y):
self.x = x
self.y = y
def __hash__(self):
return hash(self.x)
def __lt__(self, other):
try:
return self.x < other.x
except AttributeError:
return NotImplemented
assertTotallyOrdered will check that instances can be ordered correctly.
For example,
self.assertTotallyOrdered(
[None], # None should come before everything else.
[1], # Integers sort earlier.
[A(1, 'a')],
[A(2, 'b')], # 2 is after 1.
[A(3, 'c'), A(3, 'd')], # The second argument is irrelevant.
[A(4, 'z')],
['foo']) # Strings sort last.
##### Args:
* <b>`*groups`</b>: A list of groups of elements. Each group of elements is a list
of objects that are equal. The elements in each group must be less than
the elements in the group after it. For example, these groups are
totally ordered: [None], [1], [2, 2], [3].
* <b>`**kwargs`</b>: optional msg keyword argument can be passed.
- - -
#### `tf.test.TestCase.assertTrue(expr, msg=None)` {#TestCase.assertTrue}
@ -632,6 +1098,13 @@ A tuple-specific equality assertion.
differences.
- - -
#### `tf.test.TestCase.assertUrlEqual(a, b, msg=None)` {#TestCase.assertUrlEqual}
Asserts that urls are equal, ignoring ordering of query params.
- - -
#### `tf.test.TestCase.assert_(expr, msg=None)` {#TestCase.assert_}
@ -693,9 +1166,9 @@ tearDown.
- - -
#### `tf.test.TestCase.fail(msg=None)` {#TestCase.fail}
#### `tf.test.TestCase.fail(msg=None, prefix=None)` {#TestCase.fail}
Fail immediately, with the given message.
Fail immediately with the given message, optionally prefixed.
- - -
@ -747,6 +1220,13 @@ Fail immediately, with the given message.
- - -
#### `tf.test.TestCase.getRecordedProperties()` {#TestCase.getRecordedProperties}
Return any properties that the user has recorded.
- - -
#### `tf.test.TestCase.get_temp_dir()` {#TestCase.get_temp_dir}
@ -761,6 +1241,20 @@ Fail immediately, with the given message.
- - -
#### `tf.test.TestCase.recordProperty(property_name, property_value)` {#TestCase.recordProperty}
Record an arbitrary property for later use.
##### Args:
* <b>`property_name`</b>: str, name of property to record; must be a valid XML
attribute name
* <b>`property_value`</b>: value of property; must be valid XML attribute value
- - -
#### `tf.test.TestCase.run(result=None)` {#TestCase.run}
@ -786,11 +1280,18 @@ Hook method for setting up class fixture before running tests in the class.
#### `tf.test.TestCase.shortDescription()` {#TestCase.shortDescription}
Returns a one-line description of the test, or None if no
description has been provided.
Format both the test method name and the first line of its docstring.
The default implementation of this method returns the first line of
the specified test method's docstring.
If no docstring is given, only returns the method name.
This method overrides unittest.TestCase.shortDescription(), which
only returns the first line of the docstring, obscuring the name
of the test upon failure.
##### Returns:
* <b>`desc`</b>: A short description of a test method.
- - -

10
tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.learn.Evaluable.md
View File

@ -20,12 +20,14 @@ for which this evaluation was performed.
##### Args:
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] containing the input samples
for fitting the model. Can be iterator that returns arrays of features.
If set, `input_fn` must be `None`.
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] or dictionary of many matrices
containing the input samples for fitting the model. Can be iterator that returns
arrays of features or dictionary of array of features. If set, `input_fn` must
be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs] containing the
label values (class labels in classification, real numbers in
regression). Can be iterator that returns array of labels. If set,
regression) or dictionary of multiple vectors/matrices. Can be iterator
that returns array of targets or dictionary of array of targets. If set,
`input_fn` must be `None`. Note: For classification, label values must
be integers representing the class index (i.e. values from 0 to
n_classes-1).

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.ctc_beam_search_decoder.md
View File

@ -3,8 +3,8 @@
Performs beam search decoding on the logits given in input.
**Note** The `ctc_greedy_decoder` is a special case of the
`ctc_beam_search_decoder` with `top_paths=1` (but that decoder is faster
for this special case).
`ctc_beam_search_decoder` with `top_paths=1` and `beam_width=1` (but
that decoder is faster for this special case).
If `merge_repeated` is `True`, merge repeated classes in the output beams.
This means that if consecutive entries in a beam are the same,

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard4/tf.summary.SummaryDescription.RegisterExtension.md
View File

@ -1,4 +0,0 @@
#### `tf.summary.SummaryDescription.RegisterExtension(extension_handle)` {#SummaryDescription.RegisterExtension}

26
tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.image.decode_image.md Normal file
View File

@ -0,0 +1,26 @@
### `tf.image.decode_image(contents, channels=None, name=None)` {#decode_image}
Convenience function for `decode_gif`, `decode_jpeg`, and `decode_png`.
Detects whether an image is a GIF, JPEG, or PNG, and performs the appropriate
operation to convert the input bytes `string` into a `Tensor` of type `uint8`.
Note: `decode_gif` returns a 4-D array `[num_frames, height, width, 3]`, as
opposed to `decode_jpeg` and `decode_png`, which return 3-D arrays
`[height, width, num_channels]`. Make sure to take this into account when
constructing your graph if you are intermixing GIF files with JPEG and/or PNG
files.
##### Args:
* <b>`contents`</b>: 0-D `string`. The encoded image bytes.
* <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for
the decoded image.
* <b>`name`</b>: A name for the operation (optional)
##### Returns:
`Tensor` with type `uint8` with shape `[height, width, num_channels]` for
JPEG and PNG images and shape `[num_frames, height, width, 3]` for GIF
images.

2
tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.make_template.md
View File

@ -13,7 +13,7 @@ have the following properties:
that are intended to be locals can be created by specifying
`tf.Variable(..., trainable=false)`.
* The function may use variable scopes and other templates internally to
create and reuse variables, but it shouldn't use `tf.all_variables` to
create and reuse variables, but it shouldn't use `tf.global_variables` to
capture variables that are defined outside of the scope of the function.
* Internal scopes and variable names should not depend on any arguments that
are not supplied to `make_template`. In general you will get a ValueError

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard5/tf.summary.SummaryDescription.FromString.md
View File

@ -1,4 +0,0 @@
#### `tf.summary.SummaryDescription.FromString(s)` {#SummaryDescription.FromString}

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.summary.TaggedRunMetadata.RegisterExtension.md
View File

@ -1,4 +0,0 @@
#### `tf.summary.TaggedRunMetadata.RegisterExtension(extension_handle)` {#TaggedRunMetadata.RegisterExtension}

2
tensorflow/g3doc/api_docs/python/functions_and_classes/shard6/tf.train.Supervisor.md
View File

@ -174,7 +174,7 @@ Create a `Supervisor`.
ready to run the local_init_op.
The model is considered ready if it returns an empty array. Defaults to
the tensor returned from
`tf.report_uninitialized_variables(tf.all_variables())`. If `None`, the
`tf.report_uninitialized_variables(tf.global_variables())`. If `None`, the
model is not checked for readiness before running local_init_op.
* <b>`is_chief`</b>: If True, create a chief supervisor in charge of initializing
and restoring the model. If False, create a supervisor that relies

11
tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.GraphKeys.md
View File

@ -42,14 +42,3 @@ The following standard keys are defined:
* `WEIGHTS`: weights inside neural network layers
* `BIASES`: biases inside neural network layers
* `ACTIVATIONS`: activations of neural network layers
- - -
#### `tf.GraphKeys.VARIABLES` {#GraphKeys.VARIABLES}
DEPRECATED FUNCTION
THIS FUNCTION IS DEPRECATED. It will be removed after 2017-03-02.
Instructions for updating:
VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard8/tf.Variable.md
View File

@ -51,7 +51,7 @@ with tf.Session() as sess:
```
The most common initialization pattern is to use the convenience function
`global_variable_initializers()` to add an Op to the graph that initializes
`global_variables_initializer()` to add an Op to the graph that initializes
all the variables. You then run that Op after launching the graph.
```python
@ -305,7 +305,7 @@ more information on launching a graph and on sessions.
```python
v = tf.Variable([1, 2])
init = tf.global_variable_initializers()
init = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init)

16
tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.learn.Trainable.md
View File

@ -8,17 +8,17 @@ Trains a model given training data `x` predictions and `y` labels.
##### Args:
* <b>`x`</b>: Matrix of shape [n_samples, n_features...]. Can be iterator that
returns arrays of features. The training input samples for fitting the
model. If set, `input_fn` must be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs]. Can be
iterator that returns array of labels. The training label values
(class labels in classification, real numbers in regression). If set,
`input_fn` must be `None`. Note: For classification, label values must
* <b>`x`</b>: Matrix of shape [n_samples, n_features...] or the dictionary of Matrices.
Can be iterator that returns arrays of features or dictionary of arrays of features.
The training input samples for fitting the model. If set, `input_fn` must be `None`.
* <b>`y`</b>: Vector or matrix [n_samples] or [n_samples, n_outputs] or the dictionary of same.
Can be iterator that returns array of labels or dictionary of array of labels.
The training label values (class labels in classification, real numbers in regression).
If set, `input_fn` must be `None`. Note: For classification, label values must
be integers representing the class index (i.e. values from 0 to
n_classes-1).
* <b>`input_fn`</b>: Input function returning a tuple of:
features - Dictionary of string feature name to `Tensor` or `Tensor`.
features - `Tensor` or dictionary of string feature name to `Tensor`.
labels - `Tensor` or dictionary of `Tensor` with labels.
If input_fn is set, `x`, `y`, and `batch_size` must be `None`.
* <b>`steps`</b>: Number of steps for which to train model. If `None`, train forever.

4
tensorflow/g3doc/api_docs/python/functions_and_classes/shard9/tf.summary.TaggedRunMetadata.FromString.md
View File

@ -1,4 +0,0 @@
#### `tf.summary.TaggedRunMetadata.FromString(s)` {#TaggedRunMetadata.FromString}

30
tensorflow/g3doc/api_docs/python/image.md
View File

@ -216,6 +216,36 @@ the smallest output, but is slower.
- - -
### `tf.image.decode_image(contents, channels=None, name=None)` {#decode_image}
Convenience function for `decode_gif`, `decode_jpeg`, and `decode_png`.
Detects whether an image is a GIF, JPEG, or PNG, and performs the appropriate
operation to convert the input bytes `string` into a `Tensor` of type `uint8`.
Note: `decode_gif` returns a 4-D array `[num_frames, height, width, 3]`, as
opposed to `decode_jpeg` and `decode_png`, which return 3-D arrays
`[height, width, num_channels]`. Make sure to take this into account when
constructing your graph if you are intermixing GIF files with JPEG and/or PNG
files.
##### Args:
* <b>`contents`</b>: 0-D `string`. The encoded image bytes.
* <b>`channels`</b>: An optional `int`. Defaults to `0`. Number of color channels for
the decoded image.
* <b>`name`</b>: A name for the operation (optional)
##### Returns:
`Tensor` with type `uint8` with shape `[height, width, num_channels]` for
JPEG and PNG images and shape `[num_frames, height, width, 3]` for GIF
images.
## Resizing
The resizing Ops accept input images as tensors of several types. They always

1
tensorflow/g3doc/api_docs/python/index.md
View File

@ -381,6 +381,7 @@
* [`crop_and_resize`](../../api_docs/python/image.md#crop_and_resize)
* [`crop_to_bounding_box`](../../api_docs/python/image.md#crop_to_bounding_box)
* [`decode_gif`](../../api_docs/python/image.md#decode_gif)
* [`decode_image`](../../api_docs/python/image.md#decode_image)
* [`decode_jpeg`](../../api_docs/python/image.md#decode_jpeg)
* [`decode_png`](../../api_docs/python/image.md#decode_png)
* [`draw_bounding_boxes`](../../api_docs/python/image.md#draw_bounding_boxes)

16
tensorflow/g3doc/api_docs/python/nn.md
View File

@ -2469,10 +2469,11 @@ tensor. The returned tensor has shape `shape(ids) + shape(params)[1:]`.
##### Args:
* <b>`params`</b>: A list of tensors with the same type and which can be concatenated
along dimension 0. Alternatively, a `PartitionedVariable`, created by
partitioning along dimension 0. Each element must be appropriately sized
for the given `partition_strategy`.
* <b>`params`</b>: A single tensor representing the complete embedding tensor,
or a list of P tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
* <b>`ids`</b>: A `Tensor` with type `int32` or `int64` containing the ids to be looked
up in `params`.
* <b>`partition_strategy`</b>: A string specifying the partitioning strategy, relevant
@ -2512,7 +2513,8 @@ is the sum of the size of params along dimension 0.
* <b>`params`</b>: A single tensor representing the complete embedding tensor,
or a list of P tensors all of same shape except for the first dimension,
representing sharded embedding tensors. Alternatively, a
`PartitionedVariable`, created by partitioning along dimension 0.
`PartitionedVariable`, created by partitioning along dimension 0. Each
element must be appropriately sized for the given `partition_strategy`.
* <b>`sp_ids`</b>: N x M SparseTensor of int64 ids (typically from FeatureValueToId),
where N is typically batch size and M is arbitrary.
* <b>`sp_weights`</b>: either a SparseTensor of float / double weights, or None to
@ -3257,8 +3259,8 @@ is the blank label) becomes
Performs beam search decoding on the logits given in input.
**Note** The `ctc_greedy_decoder` is a special case of the
`ctc_beam_search_decoder` with `top_paths=1` (but that decoder is faster
for this special case).
`ctc_beam_search_decoder` with `top_paths=1` and `beam_width=1` (but
that decoder is faster for this special case).
If `merge_repeated` is `True`, merge repeated classes in the output beams.
This means that if consecutive entries in a beam are the same,

8
tensorflow/g3doc/api_docs/python/state_ops.md
View File

@ -66,7 +66,7 @@ with tf.Session() as sess:
```
The most common initialization pattern is to use the convenience function
`global_variable_initializers()` to add an Op to the graph that initializes
`global_variables_initializer()` to add an Op to the graph that initializes
all the variables. You then run that Op after launching the graph.
```python
@ -320,7 +320,7 @@ more information on launching a graph and on sessions.
```python
v = tf.Variable([1, 2])
init = tf.global_variable_initializers()
init = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init)
@ -2255,7 +2255,7 @@ have the following properties:
that are intended to be locals can be created by specifying
`tf.Variable(..., trainable=false)`.
* The function may use variable scopes and other templates internally to
create and reuse variables, but it shouldn't use `tf.all_variables` to
create and reuse variables, but it shouldn't use `tf.global_variables` to
capture variables that are defined outside of the scope of the function.
* Internal scopes and variable names should not depend on any arguments that
are not supplied to `make_template`. In general you will get a ValueError
@ -2739,7 +2739,7 @@ This operation outputs `ref` after the update is done.
This makes it easier to chain operations that need to use the reset value.
If values in `ref` is to be updated more than once, because there are
duplicate entires in `indices`, the order at which the updates happen
duplicate entries in `indices`, the order at which the updates happen
for each value is undefined.
Requires `updates.shape = indices.shape + ref.shape[1:]`.

481
tensorflow/g3doc/api_docs/python/summary.md
View File

@ -485,187 +485,6 @@ metadata is stored in its NodeDef. This method retrieves the description.
### `class tf.summary.SummaryDescription` {#SummaryDescription}
- - -
#### `tf.summary.SummaryDescription.ByteSize()` {#SummaryDescription.ByteSize}
- - -
#### `tf.summary.SummaryDescription.Clear()` {#SummaryDescription.Clear}
- - -
#### `tf.summary.SummaryDescription.ClearExtension(extension_handle)` {#SummaryDescription.ClearExtension}
- - -
#### `tf.summary.SummaryDescription.ClearField(field_name)` {#SummaryDescription.ClearField}
- - -
#### `tf.summary.SummaryDescription.CopyFrom(other_msg)` {#SummaryDescription.CopyFrom}
Copies the content of the specified message into the current message.
The method clears the current message and then merges the specified
message using MergeFrom.
##### Args:
* <b>`other_msg`</b>: Message to copy into the current one.
- - -
#### `tf.summary.SummaryDescription.DiscardUnknownFields()` {#SummaryDescription.DiscardUnknownFields}
- - -
#### `tf.summary.SummaryDescription.FindInitializationErrors()` {#SummaryDescription.FindInitializationErrors}
Finds required fields which are not initialized.
##### Returns:
A list of strings. Each string is a path to an uninitialized field from
the top-level message, e.g. "foo.bar[5].baz".
- - -
#### `tf.summary.SummaryDescription.FromString(s)` {#SummaryDescription.FromString}
- - -
#### `tf.summary.SummaryDescription.HasExtension(extension_handle)` {#SummaryDescription.HasExtension}
- - -
#### `tf.summary.SummaryDescription.HasField(field_name)` {#SummaryDescription.HasField}
- - -
#### `tf.summary.SummaryDescription.IsInitialized(errors=None)` {#SummaryDescription.IsInitialized}
Checks if all required fields of a message are set.
##### Args:
* <b>`errors`</b>: A list which, if provided, will be populated with the field
paths of all missing required fields.
##### Returns:
True iff the specified message has all required fields set.
- - -
#### `tf.summary.SummaryDescription.ListFields()` {#SummaryDescription.ListFields}
- - -
#### `tf.summary.SummaryDescription.MergeFrom(msg)` {#SummaryDescription.MergeFrom}
- - -
#### `tf.summary.SummaryDescription.MergeFromString(serialized)` {#SummaryDescription.MergeFromString}
- - -
#### `tf.summary.SummaryDescription.ParseFromString(serialized)` {#SummaryDescription.ParseFromString}
Parse serialized protocol buffer data into this message.
Like MergeFromString(), except we clear the object first and
do not return the value that MergeFromString returns.
- - -
#### `tf.summary.SummaryDescription.RegisterExtension(extension_handle)` {#SummaryDescription.RegisterExtension}
- - -
#### `tf.summary.SummaryDescription.SerializePartialToString()` {#SummaryDescription.SerializePartialToString}
- - -
#### `tf.summary.SummaryDescription.SerializeToString()` {#SummaryDescription.SerializeToString}
- - -
#### `tf.summary.SummaryDescription.SetInParent()` {#SummaryDescription.SetInParent}
Sets the _cached_byte_size_dirty bit to true,
and propagates this to our listener iff this was a state change.
- - -
#### `tf.summary.SummaryDescription.WhichOneof(oneof_name)` {#SummaryDescription.WhichOneof}
Returns the name of the currently set field inside a oneof, or None.
- - -
#### `tf.summary.SummaryDescription.__deepcopy__(memo=None)` {#SummaryDescription.__deepcopy__}
- - -
#### `tf.summary.SummaryDescription.__eq__(other)` {#SummaryDescription.__eq__}
- - -
#### `tf.summary.SummaryDescription.__getstate__()` {#SummaryDescription.__getstate__}
@ -673,249 +492,12 @@ Returns the name of the currently set field inside a oneof, or None.
Support the pickle protocol.
- - -
#### `tf.summary.SummaryDescription.__hash__()` {#SummaryDescription.__hash__}
- - -
#### `tf.summary.SummaryDescription.__init__(**kwargs)` {#SummaryDescription.__init__}
- - -
#### `tf.summary.SummaryDescription.__ne__(other_msg)` {#SummaryDescription.__ne__}
- - -
#### `tf.summary.SummaryDescription.__repr__()` {#SummaryDescription.__repr__}
- - -
#### `tf.summary.SummaryDescription.__setstate__(state)` {#SummaryDescription.__setstate__}
Support the pickle protocol.
- - -
#### `tf.summary.SummaryDescription.__str__()` {#SummaryDescription.__str__}
- - -
#### `tf.summary.SummaryDescription.__unicode__()` {#SummaryDescription.__unicode__}
- - -
#### `tf.summary.SummaryDescription.type_hint` {#SummaryDescription.type_hint}
Magic attribute generated for "type_hint" proto field.
- - -
### `class tf.summary.TaggedRunMetadata` {#TaggedRunMetadata}
- - -
#### `tf.summary.TaggedRunMetadata.ByteSize()` {#TaggedRunMetadata.ByteSize}
- - -
#### `tf.summary.TaggedRunMetadata.Clear()` {#TaggedRunMetadata.Clear}
- - -
#### `tf.summary.TaggedRunMetadata.ClearExtension(extension_handle)` {#TaggedRunMetadata.ClearExtension}
- - -
#### `tf.summary.TaggedRunMetadata.ClearField(field_name)` {#TaggedRunMetadata.ClearField}
- - -
#### `tf.summary.TaggedRunMetadata.CopyFrom(other_msg)` {#TaggedRunMetadata.CopyFrom}
Copies the content of the specified message into the current message.
The method clears the current message and then merges the specified
message using MergeFrom.
##### Args:
* <b>`other_msg`</b>: Message to copy into the current one.
- - -
#### `tf.summary.TaggedRunMetadata.DiscardUnknownFields()` {#TaggedRunMetadata.DiscardUnknownFields}
- - -
#### `tf.summary.TaggedRunMetadata.FindInitializationErrors()` {#TaggedRunMetadata.FindInitializationErrors}
Finds required fields which are not initialized.
##### Returns:
A list of strings. Each string is a path to an uninitialized field from
the top-level message, e.g. "foo.bar[5].baz".
- - -
#### `tf.summary.TaggedRunMetadata.FromString(s)` {#TaggedRunMetadata.FromString}
- - -
#### `tf.summary.TaggedRunMetadata.HasExtension(extension_handle)` {#TaggedRunMetadata.HasExtension}
- - -
#### `tf.summary.TaggedRunMetadata.HasField(field_name)` {#TaggedRunMetadata.HasField}
- - -
#### `tf.summary.TaggedRunMetadata.IsInitialized(errors=None)` {#TaggedRunMetadata.IsInitialized}
Checks if all required fields of a message are set.
##### Args:
* <b>`errors`</b>: A list which, if provided, will be populated with the field
paths of all missing required fields.
##### Returns:
True iff the specified message has all required fields set.
- - -
#### `tf.summary.TaggedRunMetadata.ListFields()` {#TaggedRunMetadata.ListFields}
- - -
#### `tf.summary.TaggedRunMetadata.MergeFrom(msg)` {#TaggedRunMetadata.MergeFrom}
- - -
#### `tf.summary.TaggedRunMetadata.MergeFromString(serialized)` {#TaggedRunMetadata.MergeFromString}
- - -
#### `tf.summary.TaggedRunMetadata.ParseFromString(serialized)` {#TaggedRunMetadata.ParseFromString}
Parse serialized protocol buffer data into this message.
Like MergeFromString(), except we clear the object first and
do not return the value that MergeFromString returns.
- - -
#### `tf.summary.TaggedRunMetadata.RegisterExtension(extension_handle)` {#TaggedRunMetadata.RegisterExtension}
- - -
#### `tf.summary.TaggedRunMetadata.SerializePartialToString()` {#TaggedRunMetadata.SerializePartialToString}
- - -
#### `tf.summary.TaggedRunMetadata.SerializeToString()` {#TaggedRunMetadata.SerializeToString}
- - -
#### `tf.summary.TaggedRunMetadata.SetInParent()` {#TaggedRunMetadata.SetInParent}
Sets the _cached_byte_size_dirty bit to true,
and propagates this to our listener iff this was a state change.
- - -
#### `tf.summary.TaggedRunMetadata.WhichOneof(oneof_name)` {#TaggedRunMetadata.WhichOneof}
Returns the name of the currently set field inside a oneof, or None.
- - -
#### `tf.summary.TaggedRunMetadata.__deepcopy__(memo=None)` {#TaggedRunMetadata.__deepcopy__}
- - -
#### `tf.summary.TaggedRunMetadata.__eq__(other)` {#TaggedRunMetadata.__eq__}
- - -
#### `tf.summary.TaggedRunMetadata.__getstate__()` {#TaggedRunMetadata.__getstate__}
@ -923,67 +505,4 @@ Returns the name of the currently set field inside a oneof, or None.
Support the pickle protocol.
- - -
#### `tf.summary.TaggedRunMetadata.__hash__()` {#TaggedRunMetadata.__hash__}
- - -
#### `tf.summary.TaggedRunMetadata.__init__(**kwargs)` {#TaggedRunMetadata.__init__}
- - -
#### `tf.summary.TaggedRunMetadata.__ne__(other_msg)` {#TaggedRunMetadata.__ne__}
- - -
#### `tf.summary.TaggedRunMetadata.__repr__()` {#TaggedRunMetadata.__repr__}
- - -
#### `tf.summary.TaggedRunMetadata.__setstate__(state)` {#TaggedRunMetadata.__setstate__}
Support the pickle protocol.
- - -
#### `tf.summary.TaggedRunMetadata.__str__()` {#TaggedRunMetadata.__str__}
- - -
#### `tf.summary.TaggedRunMetadata.__unicode__()` {#TaggedRunMetadata.__unicode__}
- - -
#### `tf.summary.TaggedRunMetadata.run_metadata` {#TaggedRunMetadata.run_metadata}
Magic attribute generated for "run_metadata" proto field.
- - -
#### `tf.summary.TaggedRunMetadata.tag` {#TaggedRunMetadata.tag}
Magic attribute generated for "tag" proto field.

521
tensorflow/g3doc/api_docs/python/test.md
View File

@ -213,6 +213,125 @@ Checks that for all elements of farray1 and farray2
* <b>`err`</b>: a float value.
- - -
#### `tf.test.TestCase.assertBetween(value, minv, maxv, msg=None)` {#TestCase.assertBetween}
Asserts that value is between minv and maxv (inclusive).
- - -
#### `tf.test.TestCase.assertCommandFails(command, regexes, env=None, close_fds=True, msg=None)` {#TestCase.assertCommandFails}
Asserts a shell command fails and the error matches a regex in a list.
##### Args:
* <b>`command`</b>: List or string representing the command to run.
* <b>`regexes`</b>: the list of regular expression strings.
* <b>`env`</b>: Dictionary of environment variable settings.
* <b>`close_fds`</b>: Whether or not to close all open fd's in the child after
forking.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertCommandSucceeds(command, regexes=('',), env=None, close_fds=True, msg=None)` {#TestCase.assertCommandSucceeds}
Asserts that a shell command succeeds (i.e. exits with code 0).
##### Args:
* <b>`command`</b>: List or string representing the command to run.
* <b>`regexes`</b>: List of regular expression byte strings that match success.
* <b>`env`</b>: Dictionary of environment variable settings.
* <b>`close_fds`</b>: Whether or not to close all open fd's in the child after
forking.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsExactSubsequence(container, subsequence, msg=None)` {#TestCase.assertContainsExactSubsequence}
Assert that "container" contains "subsequence" as an exact subsequence.
Asserts that "container" contains all the elements of "subsequence", in
order, and without other elements interspersed. For example, [1, 2, 3] is an
exact subsequence of [0, 0, 1, 2, 3, 0] but not of [0, 0, 1, 2, 0, 3, 0].
##### Args:
* <b>`container`</b>: the list we're testing for subsequence inclusion.
* <b>`subsequence`</b>: the list we hope will be an exact subsequence of container.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsInOrder(strings, target, msg=None)` {#TestCase.assertContainsInOrder}
Asserts that the strings provided are found in the target in order.
This may be useful for checking HTML output.
##### Args:
* <b>`strings`</b>: A list of strings, such as [ 'fox', 'dog' ]
* <b>`target`</b>: A target string in which to look for the strings, such as
'The quick brown fox jumped over the lazy dog'.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsSubsequence(container, subsequence, msg=None)` {#TestCase.assertContainsSubsequence}
Assert that "container" contains "subsequence" as a subsequence.
Asserts that "container" contains all the elements of "subsequence", in
order, but possibly with other elements interspersed. For example, [1, 2, 3]
is a subsequence of [0, 0, 1, 2, 0, 3, 0] but not of [0, 0, 1, 3, 0, 2, 0].
##### Args:
* <b>`container`</b>: the list we're testing for subsequence inclusion.
* <b>`subsequence`</b>: the list we hope will be a subsequence of container.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertContainsSubset(expected_subset, actual_set, msg=None)` {#TestCase.assertContainsSubset}
Checks whether actual iterable is a superset of expected iterable.
- - -
#### `tf.test.TestCase.assertCountEqual(*args, **kwargs)` {#TestCase.assertCountEqual}
An unordered sequence specific comparison.
Equivalent to assertItemsEqual(). This method is a compatibility layer
for Python 3k, since 2to3 does not convert assertItemsEqual() calls into
assertCountEqual() calls.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertDeviceEqual(device1, device2)` {#TestCase.assertDeviceEqual}
@ -235,9 +354,48 @@ Checks whether actual is a superset of expected.
- - -
#### `tf.test.TestCase.assertDictEqual(d1, d2, msg=None)` {#TestCase.assertDictEqual}
#### `tf.test.TestCase.assertDictEqual(a, b, msg=None)` {#TestCase.assertDictEqual}
Raises AssertionError if a and b are not equal dictionaries.
##### Args:
* <b>`a`</b>: A dict, the expected value.
* <b>`b`</b>: A dict, the actual value.
* <b>`msg`</b>: An optional str, the associated message.
##### Raises:
* <b>`AssertionError`</b>: if the dictionaries are not equal.
- - -
#### `tf.test.TestCase.assertEmpty(container, msg=None)` {#TestCase.assertEmpty}
Assert that an object has zero length.
##### Args:
* <b>`container`</b>: Anything that implements the collections.Sized interface.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertEndsWith(actual, expected_end, msg=None)` {#TestCase.assertEndsWith}
Assert that actual.endswith(expected_end) is True.
##### Args:
* <b>`actual`</b>: str
* <b>`expected_end`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
@ -322,10 +480,11 @@ Included for symmetry with assertIsNone.
- - -
#### `tf.test.TestCase.assertItemsEqual(expected_seq, actual_seq, msg=None)` {#TestCase.assertItemsEqual}
#### `tf.test.TestCase.assertItemsEqual(*args, **kwargs)` {#TestCase.assertItemsEqual}
An unordered sequence specific comparison. It asserts that
actual_seq and expected_seq have the same element counts.
An unordered sequence specific comparison.
It asserts that actual_seq and expected_seq have the same element counts.
Equivalent to::
self.assertEqual(Counter(iter(actual_seq)),
@ -338,6 +497,30 @@ Asserts that each element has the same count in both sequences.
- [0, 1, 1] and [1, 0, 1] compare equal.
- [0, 0, 1] and [0, 1] compare unequal.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertJsonEqual(first, second, msg=None)` {#TestCase.assertJsonEqual}
Asserts that the JSON objects defined in two strings are equal.
A summary of the differences will be included in the failure message
using assertSameStructure.
##### Args:
* <b>`first`</b>: A string contining JSON to decode and compare to second.
* <b>`second`</b>: A string contining JSON to decode and compare to first.
* <b>`msg`</b>: Additional text to include in the failure message.
- - -
@ -407,6 +590,13 @@ if not.
* <b>`msg`</b>: An optional string message to append to the failure message.
- - -
#### `tf.test.TestCase.assertNoCommonElements(expected_seq, actual_seq, msg=None)` {#TestCase.assertNoCommonElements}
Checks whether actual iterable and expected iterable are disjoint.
- - -
#### `tf.test.TestCase.assertNotAlmostEqual(first, second, places=None, msg=None, delta=None)` {#TestCase.assertNotAlmostEqual}
@ -437,6 +627,33 @@ as significant digits (measured from the most signficant digit).
Objects that are equal automatically fail.
- - -
#### `tf.test.TestCase.assertNotEmpty(container, msg=None)` {#TestCase.assertNotEmpty}
Assert that an object has non-zero length.
##### Args:
* <b>`container`</b>: Anything that implements the collections.Sized interface.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertNotEndsWith(actual, unexpected_end, msg=None)` {#TestCase.assertNotEndsWith}
Assert that actual.endswith(unexpected_end) is False.
##### Args:
* <b>`actual`</b>: str
* <b>`unexpected_end`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertNotEqual(first, second, msg=None)` {#TestCase.assertNotEqual}
@ -474,6 +691,20 @@ Included for symmetry with assertIsInstance.
Fail the test if the text matches the regular expression.
- - -
#### `tf.test.TestCase.assertNotStartsWith(actual, unexpected_start, msg=None)` {#TestCase.assertNotStartsWith}
Assert that actual.startswith(unexpected_start) is False.
##### Args:
* <b>`actual`</b>: str
* <b>`unexpected_start`</b>: str
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertProtoEquals(expected_message_maybe_ascii, message)` {#TestCase.assertProtoEquals}
@ -548,6 +779,38 @@ Asserts that the message in a raised exception matches a regexp.
* <b>`kwargs`</b>: Extra kwargs.
- - -
#### `tf.test.TestCase.assertRaisesWithLiteralMatch(expected_exception, expected_exception_message, callable_obj=None, *args, **kwargs)` {#TestCase.assertRaisesWithLiteralMatch}
Asserts that the message in a raised exception equals the given string.
Unlike assertRaisesRegexp, this method takes a literal string, not
a regular expression.
with self.assertRaisesWithLiteralMatch(ExType, 'message'):
DoSomething()
##### Args:
* <b>`expected_exception`</b>: Exception class expected to be raised.
* <b>`expected_exception_message`</b>: String message expected in the raised
exception. For a raise exception e, expected_exception_message must
equal str(e).
* <b>`callable_obj`</b>: Function to be called, or None to return a context.
* <b>`args`</b>: Extra args.
* <b>`kwargs`</b>: Extra kwargs.
##### Returns:
A context manager if callable_obj is None. Otherwise, None.
##### Raises:
self.failureException if callable_obj does not raise a macthing exception.
- - -
#### `tf.test.TestCase.assertRaisesWithPredicateMatch(exception_type, expected_err_re_or_predicate)` {#TestCase.assertRaisesWithPredicateMatch}
@ -572,6 +835,71 @@ predicate search.
exception.
- - -
#### `tf.test.TestCase.assertRaisesWithRegexpMatch(expected_exception, expected_regexp, callable_obj=None, *args, **kwargs)` {#TestCase.assertRaisesWithRegexpMatch}
Asserts that the message in a raised exception matches the given regexp.
This is just a wrapper around assertRaisesRegexp. Please use
assertRaisesRegexp instead of assertRaisesWithRegexpMatch.
##### Args:
* <b>`expected_exception`</b>: Exception class expected to be raised.
* <b>`expected_regexp`</b>: Regexp (re pattern object or string) expected to be
found in error message.
* <b>`callable_obj`</b>: Function to be called, or None to return a context.
* <b>`args`</b>: Extra args.
* <b>`kwargs`</b>: Extra keyword args.
##### Returns:
A context manager if callable_obj is None. Otherwise, None.
##### Raises:
self.failureException if callable_obj does not raise a macthing exception.
- - -
#### `tf.test.TestCase.assertRegexMatch(actual_str, regexes, message=None)` {#TestCase.assertRegexMatch}
Asserts that at least one regex in regexes matches str.
If possible you should use assertRegexpMatches, which is a simpler
version of this method. assertRegexpMatches takes a single regular
expression (a string or re compiled object) instead of a list.
Notes:
1. This function uses substring matching, i.e. the matching
succeeds if *any* substring of the error message matches *any*
regex in the list. This is more convenient for the user than
full-string matching.
2. If regexes is the empty list, the matching will always fail.
3. Use regexes=[''] for a regex that will always pass.
4. '.' matches any single character *except* the newline. To
match any character, use '(.|
)'.
5. '^' matches the beginning of each line, not just the beginning
of the string. Similarly, '$' matches the end of each line.
6. An exception will be thrown if regexes contains an invalid
regex.
Args:
actual_str: The string we try to match with the items in regexes.
regexes: The regular expressions we want to match against str.
See "Notes" above for detailed notes on how this is interpreted.
message: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertRegexpMatches(text, expected_regexp, msg=None)` {#TestCase.assertRegexpMatches}
@ -579,6 +907,79 @@ predicate search.
Fail the test unless the text matches the regular expression.
- - -
#### `tf.test.TestCase.assertSameElements(expected_seq, actual_seq, msg=None)` {#TestCase.assertSameElements}
Assert that two sequences have the same elements (in any order).
This method, unlike assertItemsEqual, doesn't care about any
duplicates in the expected and actual sequences.
>> assertSameElements([1, 1, 1, 0, 0, 0], [0, 1])
# Doesn't raise an AssertionError
If possible, you should use assertItemsEqual instead of
assertSameElements.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`msg`</b>: The message to be printed if the test fails.
- - -
#### `tf.test.TestCase.assertSameStructure(a, b, aname='a', bname='b', msg=None)` {#TestCase.assertSameStructure}
Asserts that two values contain the same structural content.
The two arguments should be data trees consisting of trees of dicts and
lists. They will be deeply compared by walking into the contents of dicts
and lists; other items will be compared using the == operator.
If the two structures differ in content, the failure message will indicate
the location within the structures where the first difference is found.
This may be helpful when comparing large structures.
##### Args:
* <b>`a`</b>: The first structure to compare.
* <b>`b`</b>: The second structure to compare.
* <b>`aname`</b>: Variable name to use for the first structure in assertion messages.
* <b>`bname`</b>: Variable name to use for the second structure.
* <b>`msg`</b>: Additional text to include in the failure message.
- - -
#### `tf.test.TestCase.assertSequenceAlmostEqual(expected_seq, actual_seq, places=None, msg=None, delta=None)` {#TestCase.assertSequenceAlmostEqual}
An approximate equality assertion for ordered sequences.
Fail if the two sequences are unequal as determined by their value
differences rounded to the given number of decimal places (default 7) and
comparing to zero, or by comparing that the difference between each value
in the two sequences is more than the given delta.
Note that decimal places (from zero) are usually not the same as significant
digits (measured from the most signficant digit).
If the two sequences compare equal then they will automatically compare
almost equal.
##### Args:
* <b>`expected_seq`</b>: A sequence containing elements we are expecting.
* <b>`actual_seq`</b>: The sequence that we are testing.
* <b>`places`</b>: The number of decimal places to compare.
* <b>`msg`</b>: The message to be printed if the test fails.
* <b>`delta`</b>: The OK difference between compared values.
- - -
#### `tf.test.TestCase.assertSequenceEqual(seq1, seq2, msg=None, seq_type=None)` {#TestCase.assertSequenceEqual}
@ -599,6 +1000,26 @@ which can be indexed, has a length, and has an equality operator.
differences.
- - -
#### `tf.test.TestCase.assertSequenceStartsWith(prefix, whole, msg=None)` {#TestCase.assertSequenceStartsWith}
An equality assertion for the beginning of ordered sequences.
If prefix is an empty sequence, it will raise an error unless whole is also
an empty sequence.
If prefix is not a sequence, it will raise an error if the first element of
whole does not match.
##### Args:
* <b>`prefix`</b>: A sequence expected at the beginning of the whole parameter.
* <b>`whole`</b>: The sequence in which to look for prefix.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertSetEqual(set1, set2, msg=None)` {#TestCase.assertSetEqual}
@ -650,6 +1071,51 @@ Assert that actual.startswith(expected_start) is True.
* <b>`msg`</b>: Optional message to report on failure.
- - -
#### `tf.test.TestCase.assertTotallyOrdered(*groups, **kwargs)` {#TestCase.assertTotallyOrdered}
Asserts that total ordering has been implemented correctly.
For example, say you have a class A that compares only on its attribute x.
Comparators other than __lt__ are omitted for brevity.
class A(object):
def __init__(self, x, y):
self.x = x
self.y = y
def __hash__(self):
return hash(self.x)
def __lt__(self, other):
try:
return self.x < other.x
except AttributeError:
return NotImplemented
assertTotallyOrdered will check that instances can be ordered correctly.
For example,
self.assertTotallyOrdered(
[None], # None should come before everything else.
[1], # Integers sort earlier.
[A(1, 'a')],
[A(2, 'b')], # 2 is after 1.
[A(3, 'c'), A(3, 'd')], # The second argument is irrelevant.
[A(4, 'z')],
['foo']) # Strings sort last.
##### Args:
* <b>`*groups`</b>: A list of groups of elements. Each group of elements is a list
of objects that are equal. The elements in each group must be less than
the elements in the group after it. For example, these groups are
totally ordered: [None], [1], [2, 2], [3].
* <b>`**kwargs`</b>: optional msg keyword argument can be passed.
- - -
#### `tf.test.TestCase.assertTrue(expr, msg=None)` {#TestCase.assertTrue}
@ -672,6 +1138,13 @@ A tuple-specific equality assertion.
differences.
- - -
#### `tf.test.TestCase.assertUrlEqual(a, b, msg=None)` {#TestCase.assertUrlEqual}
Asserts that urls are equal, ignoring ordering of query params.
- - -
#### `tf.test.TestCase.assert_(expr, msg=None)` {#TestCase.assert_}
@ -733,9 +1206,9 @@ tearDown.
- - -
#### `tf.test.TestCase.fail(msg=None)` {#TestCase.fail}
#### `tf.test.TestCase.fail(msg=None, prefix=None)` {#TestCase.fail}
Fail immediately, with the given message.
Fail immediately with the given message, optionally prefixed.
- - -
@ -787,6 +1260,13 @@ Fail immediately, with the given message.
- - -
#### `tf.test.TestCase.getRecordedProperties()` {#TestCase.getRecordedProperties}
Return any properties that the user has recorded.
- - -
#### `tf.test.TestCase.get_temp_dir()` {#TestCase.get_temp_dir}
@ -801,6 +1281,20 @@ Fail immediately, with the given message.
- - -
#### `tf.test.TestCase.recordProperty(property_name, property_value)` {#TestCase.recordProperty}
Record an arbitrary property for later use.
##### Args:
* <b>`property_name`</b>: str, name of property to record; must be a valid XML
attribute name
* <b>`property_value`</b>: value of property; must be valid XML attribute value
- - -
#### `tf.test.TestCase.run(result=None)` {#TestCase.run}
@ -826,11 +1320,18 @@ Hook method for setting up class fixture before running tests in the class.
#### `tf.test.TestCase.shortDescription()` {#TestCase.shortDescription}
Returns a one-line description of the test, or None if no
description has been provided.
Format both the test method name and the first line of its docstring.
The default implementation of this method returns the first line of
the specified test method's docstring.
If no docstring is given, only returns the method name.
This method overrides unittest.TestCase.shortDescription(), which
only returns the first line of the docstring, obscuring the name
of the test upon failure.
##### Returns:
* <b>`desc`</b>: A short description of a test method.
- - -

4
tensorflow/g3doc/api_docs/python/train.md
View File

@ -1413,7 +1413,7 @@ debiased (see docstring in `assign_moving_average` for more details).
shadow variables are created with `trainable=False` and added to the
`GraphKeys.ALL_VARIABLES` collection. They will be returned by calls to
`tf.all_variables()`.
`tf.global_variables()`.
Returns an op that updates all shadow variables as described above.
@ -2347,7 +2347,7 @@ Create a `Supervisor`.
ready to run the local_init_op.
The model is considered ready if it returns an empty array. Defaults to
the tensor returned from
`tf.report_uninitialized_variables(tf.all_variables())`. If `None`, the
`tf.report_uninitialized_variables(tf.global_variables())`. If `None`, the
model is not checked for readiness before running local_init_op.
* <b>`is_chief`</b>: If True, create a chief supervisor in charge of initializing
and restoring the model. If False, create a supervisor that relies