975afd7fe660cbe68b0f626f16bdfcd5c2f1b383 |
|
15-Jan-2018 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Remove lenient naming in tf.Saver. PiperOrigin-RevId: 181943238
/external/tensorflow/tensorflow/python/training/saver_test.py
|
733a8ed8918a548867b110c452b67c95dda537e2 |
|
12-Jan-2018 |
Skye Wanderman-Milne <skyewm@google.com> |
Sundry small changes to enable the C API in more tests. PiperOrigin-RevId: 181668268
/external/tensorflow/tensorflow/python/training/saver_test.py
|
da1588ccf5c62eccac8013673359ac15b43eb394 |
|
03-Jan-2018 |
A. Unique TensorFlower <gardener@tensorflow.org> |
meta_graph export: Add support to strip default valued attributes. Following APIs now accept an additional argument (`strip_default_attrs`) to enable/disable (default:disabled) stripping of default valued attributes in a NodeDef: o meta_graph: export_meta_graph, create_meta_graph. o saver: Saver.save, Saver.export_meta_graph. o builder: SavedModelBuilder.add_meta_graph, SavedModelBuilder.add_meta_graph_and_variables. o estimator: Estimator.export_savedmodel. Related changes: o Pywrap C++ AreAttrValuesEqual to compare two AttrValue instances. This allows for a single/canonical way of comparing AttrValues in C++/Python. o Add a utility method to meta_graph.py to get the node def by name in a graph def. o Update SavedModelBuilder documentation on relevance of strip_default_attrs flag. PiperOrigin-RevId: 180619001
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9620b2df63854538357bf41f4d9761499e8e573d |
|
07-Dec-2017 |
Igor Saprykin <isaprykin@google.com> |
Fix the issue with shared saver on GPU. `ShardedFilename` and ``MergeV2Checkpoints/checkpoint_prefixes` operations were placed on GPU even though there are no GPU kernels for them. PiperOrigin-RevId: 178276605
/external/tensorflow/tensorflow/python/training/saver_test.py
|
af8a5507937108a41781ba117fa16edd3b1091b5 |
|
06-Dec-2017 |
Derek Murray <mrry@google.com> |
Always include the function library when exporting a MetaGraphDef. Previously, some code paths through `tf.train.export_meta_graph()` did not ensure that the function library was persisted in the resulting `MetaGraphDef`. This would break serialization for meta-graphs that included `tf.data` pipelines that used functions. This fix ensures that the library is copied to all such meta-graphs. Fixes #15019. Fixes #14143. PiperOrigin-RevId: 178033103
/external/tensorflow/tensorflow/python/training/saver_test.py
|
b2db981a6731e978453862a73dab892bc674db68 |
|
01-Dec-2017 |
Sourabh Bajaj <sourabhbajaj@google.com> |
Merge changes from github. PiperOrigin-RevId: 177526301
/external/tensorflow/tensorflow/python/training/saver_test.py
|
c81a8ae591cf43b6d10b887dfb22a780af3beec0 |
|
28-Nov-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Make sure that additional ops added by Savers to read ResourceVariables are added to the graph in a deterministic way. For ResourceVariables (op "VarHandleOp"), ops.internal_convert_to_tensor will add new ops such as "Read_8/ReadVariableOp". If op_list is cast to a set, as before this change, then adding these new ops made graph construction non-deterministic. PiperOrigin-RevId: 177185279
/external/tensorflow/tensorflow/python/training/saver_test.py
|
06674161dc5c6b860bea559c2eef0217d49d86ca |
|
25-Oct-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Variable name remapping of saver/restore. PiperOrigin-RevId: 173431453
/external/tensorflow/tensorflow/python/training/saver_test.py
|
ff0530067435fea5c51605c2e7dfd55f6fe8dfe1 |
|
20-Oct-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Avoid silent variable sharing with ResourceVariable class. PiperOrigin-RevId: 172905986
/external/tensorflow/tensorflow/python/training/saver_test.py
|
08aeb0f960329efa7f477fd184d2e676a96da415 |
|
18-Oct-2017 |
Allen Lavoie <allenl@google.com> |
Automated g4 rollback of changelist 172336111 PiperOrigin-RevId: 172645893
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0a572887ffa9879d6a303aafd6de9e288776fc8f |
|
16-Oct-2017 |
Allen Lavoie <allenl@google.com> |
Automated g4 rollback of changelist 172039259 PiperOrigin-RevId: 172336111
/external/tensorflow/tensorflow/python/training/saver_test.py
|
dec1c9296f72e146423d5cb2fffed1c65ef4e8d6 |
|
13-Oct-2017 |
Ali Yahya <alive@google.com> |
TFE: Adds a destructor to ResourceVariables in Python that destroys the underlying resource. This makes the lifetime of the underlying resource match that of its corresponding Python object. PiperOrigin-RevId: 172039259
/external/tensorflow/tensorflow/python/training/saver_test.py
|
ee50560b5fd2b1112e82377d7d094a0e6918f935 |
|
29-Sep-2017 |
Gunhan Gulsoy <gunan@google.com> |
Mock out time to avoid flakiness in saver_test. PiperOrigin-RevId: 170522593
/external/tensorflow/tensorflow/python/training/saver_test.py
|
64cca2be776a332e1e9e8e7c6bbf1b170020e819 |
|
27-Sep-2017 |
Mark Heffernan <meheff@google.com> |
Do not simplify Tuple->GetTupleElement->Tuple constructs in TupleSimplifier if the input and output tuples are not compatible. PiperOrigin-RevId: 170213262
/external/tensorflow/tensorflow/python/training/saver_test.py
|
e2b96109c25d42b362c238dc3785e38083137d07 |
|
27-Sep-2017 |
Jianwei Xie <xiejw@google.com> |
Adds implementation for tf.estimator.train_and_evaluate PiperOrigin-RevId: 170207452
/external/tensorflow/tensorflow/python/training/saver_test.py
|
3076ee0a760ec3aace7a77778951df9033103e40 |
|
27-Sep-2017 |
Martin Wicke <wicke@google.com> |
Fix flaky saver_test. Don't trust sleep to sleep through the night. PiperOrigin-RevId: 170207579
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f3f4f3fbcddf4b7d65ffe85c4e38e7a07d5f74fb |
|
15-Sep-2017 |
Ali Yahya <alive@google.com> |
When Eager Execution is enabled, TensorFlow now no longer relies on global collections to keep track of ResourceVariables. Instead, they are tracked by the user as normal Python objects. In a subsequent CL, we'll make the lifetime of a variable's underlying resource match the lifetime of the corresponding Python object. For this to happen, there must be no everlasting global Python references to said variables. More specifically, this change forces the `collections` flag in ResourceVariable's constructor to be None when Eager is enabled. It also raises an error on calls to get_collection() for variable collections. PiperOrigin-RevId: 168754146
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8cb56357eab5ecf847f3342ca154894badb4fa76 |
|
14-Sep-2017 |
Igor Saprykin <isaprykin@google.com> |
Cover CheckpointState in the tf.Saver's unit tests. We know that there is a depenedency between the `_MaybeDeleteOldCheckpoints` call and the `_update_checkpoint_state` call via `self.last_checkpoints`. Yet, when I re-arrange the lines none of tests fail. With the increased coverage, the dependency becomes explicit, which is helpful when changing the saver's implementation. PiperOrigin-RevId: 168744975
/external/tensorflow/tensorflow/python/training/saver_test.py
|
557af84389abb7d71509d4ca08e4960d7361eadd |
|
07-Sep-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
A eager Saver that allows restore on create. PiperOrigin-RevId: 167789332
/external/tensorflow/tensorflow/python/training/saver_test.py
|
4359ad606c613e6491eb9e61fc32b5a24dba4b5e |
|
26-Aug-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add support for eager mode to saver. PiperOrigin-RevId: 166545870
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8a90f56ede4c0cecedbabab01e258a275ccec183 |
|
22-Aug-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Introduces a temporary "lenient naming" mode in tf.Saver that tries to load key + ":0" whenever a key does not exist in a checkpoint. This behavior is for Google-internal use only and will be removed shortly after January 1st 2018. PiperOrigin-RevId: 166081238
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a3fa1b5b07374abc1bf335226eb56af8002a4c92 |
|
04-Aug-2017 |
Anna R <annarev@google.com> |
Automated g4 rollback of changelist 164279921 PiperOrigin-RevId: 164305108
/external/tensorflow/tensorflow/python/training/saver_test.py
|
301230bee2a2a83dac63ca6266bda5088fc45675 |
|
04-Aug-2017 |
Benoit Steiner <bsteiner@google.com> |
Avoid creating Constant nodes in cases where we know we'll end up feeding actual data. This makes debugging constant folding easier. PiperOrigin-RevId: 164279921
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0b6474d5c4abebd6ebe951acb120613a5b66ec50 |
|
24-Jun-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Avoid hardcoded names for temporary files in tests. These tests (and examples that are run as tests) were using hardcoded names for temporary files. This failed when multiple copies of these tests were run in parallel, or even successively by different users, where the second run could not overwrite files left by the first. This change uses the TEST_TMPDIR environment variable used by bazel's test runner to choose a temporary directory. If that directory is not set, /tmp is used, as before. PiperOrigin-RevId: 160026924
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9a8e455a0af340320109f7b347b2aefc0af4bd11 |
|
26-May-2017 |
David Soergel <soergel@google.com> |
Add option to strip unused SaverDefs and associated nodes during export. PiperOrigin-RevId: 157177502
/external/tensorflow/tensorflow/python/training/saver_test.py
|
33bcb53486aa286ad16b0d1d7a2715febf696364 |
|
22-May-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Allow clients to pass a filename to the constructor of Saver. PiperOrigin-RevId: 156741424
/external/tensorflow/tensorflow/python/training/saver_test.py
|
060b7e25aaebda78a858456ba37ee36b1f5dc65d |
|
16-May-2017 |
Sherry Moore <sherrym@google.com> |
Ignore TypeError when striping or prepending name scopes. PiperOrigin-RevId: 156112135
/external/tensorflow/tensorflow/python/training/saver_test.py
|
02f4722a5bcc2ce3be9025112a780d1bc08918f9 |
|
09-May-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
These checks only need to be done after the fact if an error occurs while saving. PiperOrigin-RevId: 155450121
/external/tensorflow/tensorflow/python/training/saver_test.py
|
95ca363c6caf5d152eef594c7fc3703a5d490070 |
|
27-Apr-2017 |
Vijay Vasudevan <vrv@google.com> |
Make error message when explicitly adding an invalid device more clear. Before, you would get something like: `InvalidArgumentError: Cannot assign a device to node 'save/RestoreV2_45': Could not satisfy explicit device specification '/job:ps/task:1/device:CPU:0' because no devices matching that specification are registered in this process; available devices: /job:localhost/replica:0/task:0/cpu:0 [[Node: save/RestoreV2_45 = RestoreV2[dtypes=[DT_FLOAT], _device="/job:ps/task:1/device:CPU:0"](save/Const, save/RestoreV2_45/tensor_names, save/RestoreV2_45/shape_and_slices)]] Caused by op u'save/RestoreV2_45', defined` And now this reads: `InvalidArgumentError: Cannot assign a device for operation 'save/RestoreV2_45': Operation was explicitly assigned to '/job:ps/task:1/device:CPU:0' but available devices are [ /job:localhost/replica:0/task:0/cpu:0 ]. Make sure the device specification refers to a valid device. This drops the additional debug_info because this condition is tightly scoped to lack of devices, and so node information is unnecessary. Change: 154379360
/external/tensorflow/tensorflow/python/training/saver_test.py
|
4152dfba9380f273411502d975a9c6b22ebdba42 |
|
13-Apr-2017 |
Eugene Brevdo <ebrevdo@google.com> |
Fix MetaGraphDef comparison. Don't compare the collection_def protos in serialized form; this leads to errors when the proto serialization isn't exactly identical. Change: 153105424
/external/tensorflow/tensorflow/python/training/saver_test.py
|
6e81b34a1a4cd86a406bdeb84112c54948c2b7ac |
|
04-Apr-2017 |
Alexandre Passos <apassos@google.com> |
Saving resource variables with a caching device. Change: 152171539
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0864bede9c68ae6b061adf738c433e65012402a2 |
|
03-Apr-2017 |
Benoit Steiner <bsteiner@google.com> |
Simplified the testImportIntoNamescope test Change: 152054364
/external/tensorflow/tensorflow/python/training/saver_test.py
|
2843a7867d51c2cf065b85899ea0b9564e4d9db9 |
|
31-Mar-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Adds option to Saver to write relative paths in checkpoint state file. Change: 151791689
/external/tensorflow/tensorflow/python/training/saver_test.py
|
19ef82151be358698f9ab8702ed5575afc94f110 |
|
15-Mar-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Make SavedModel exports include all the SAVEABLE objects and not just global variables. Change: 150243023
/external/tensorflow/tensorflow/python/training/saver_test.py
|
64690a46f07e1f7be4b01ba2c79900847c245778 |
|
22-Feb-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Fix export scope handling in saver. Change: 148231524
/external/tensorflow/tensorflow/python/training/saver_test.py
|
639b4e71f532761a4840b1cdbaea55ad0917c75b |
|
08-Feb-2017 |
Benoit Steiner <bsteiner@google.com> |
Merge changes from github. Change: 146918929
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0180c71dae4253474363e330ec6764fcfc72968e |
|
02-Feb-2017 |
Jonathan Hseu <jhseu@google.com> |
Fix the Windows build by skipping the saver URI test on windows. The better fix is to add Windows support to ParseURI, but that requires more significant changes, so leaving it as a TODO for now. Change: 146314223
/external/tensorflow/tensorflow/python/training/saver_test.py
|
921e4d2ac45b217ebe055546138cb139fc71ddca |
|
01-Feb-2017 |
Jonathan Hseu <jhseu@google.com> |
Add a Saver test to ensure it continues working on URIs. Change: 146179848
/external/tensorflow/tensorflow/python/training/saver_test.py
|
43c2f2f3058d56fad95aa32d0efb30b06ab24758 |
|
27-Jan-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Enables all optimizers for dense resource variables. Also fixes small bug with saving partitioned resource variables. Change: 145828125
/external/tensorflow/tensorflow/python/training/saver_test.py
|
db249414890236c92e9ac45e231bf5c26ec503ef |
|
17-Jan-2017 |
Patrick Nguyen <drpng@google.com> |
Merge changes from github. Change: 144729490
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0e226af7eed5e2764aa8acb825af4cd3e06d2452 |
|
11-Jan-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Switch tf.concat_v2 references in third_party/tensorflow to tf.concat. Change: 144153795
/external/tensorflow/tensorflow/python/training/saver_test.py
|
333dc32ff79af21484695157f3d141dc776f7c02 |
|
05-Jan-2017 |
Martin Wicke <wicke@google.com> |
Change arg order for {softmax,sparse_softmax,sigmoid}_cross_entropy_with_logits to be (labels, predictions), and force use of named args to avoid accidents. Change: 143629623
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a1e475c39440ae0352c69faf6097e5f13e7d1f37 |
|
29-Dec-2016 |
Gunhan Gulsoy <gunan@google.com> |
Fix: Create a unique temporary directory everytime a test runs. Change: 143135973
/external/tensorflow/tensorflow/python/training/saver_test.py
|
bed8383c27a0a7225e6fc7ff59a2cd6388fb4d09 |
|
23-Dec-2016 |
Jonathan Hseu <jhseu@google.com> |
Merge changes from github. Change: 142805270
/external/tensorflow/tensorflow/python/training/saver_test.py
|
ede3c12a11cd6858eef4de52b7697299743d4660 |
|
21-Dec-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Uses VariableV2 and removes shape-setting code in tf.Variable. Change: 142613993
/external/tensorflow/tensorflow/python/training/saver_test.py
|
58201a058853de647b37ddb0ccf63d89b2357f03 |
|
17-Dec-2016 |
Justine Tunney <jart@google.com> |
Remove hourglass imports from even more tests Change: 142318245
/external/tensorflow/tensorflow/python/training/saver_test.py
|
b183b6d0c8ff5ddf0cc48f21fc03341bf56c75f7 |
|
16-Dec-2016 |
Andrew Selle <aselle@google.com> |
Change remaining tf.mul -> tf.multiply, tf.neg -> tf.negative, and tf.sub -> tf.subtract tf.negative, tf.multiply, tf.subtract are the new names Change: 142257628
/external/tensorflow/tensorflow/python/training/saver_test.py
|
1cb96893a64f59b7265f9def9968f7bed1e57662 |
|
09-Dec-2016 |
Andrew Harp <andrewharp@google.com> |
Merge changes from github. Additionally: - change single quotes to double quotes to make path rewriting easier - guard windows lib reference with PLATFORM_WINDOWS - fixed failing kmeans test Change: 141515942
/external/tensorflow/tensorflow/python/training/saver_test.py
|
d4eb834824d79c6a64a3c4a1c4a88b434b73e63e |
|
07-Dec-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Switch all tf.concat(concat_dim, value, name) calls in third_party/tensorflow to tf.concat_v2(value, axis, name). Change: 141255675
/external/tensorflow/tensorflow/python/training/saver_test.py
|
6de612e7767c4114d65cb06fc6606436e3081dc3 |
|
06-Dec-2016 |
Andrew Selle <aselle@google.com> |
Force fix documentation causing open source blocks. Change: 141201358
/external/tensorflow/tensorflow/python/training/saver_test.py
|
01b1ae3f3af381e5f26929815f78ab5982432cf8 |
|
02-Dec-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Make Saver support passing PartitionedVariables in the list/dict of specific variables to save or restore. Change: 140788798
/external/tensorflow/tensorflow/python/training/saver_test.py
|
aa4aadef0492005f60c03aa40789efad719752f0 |
|
23-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Makes resource variables saveable/restorable. Change: 140055398
/external/tensorflow/tensorflow/python/training/saver_test.py
|
5da8731b128aed7f8434750984d30ae6b1b0e95b |
|
17-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Adding __version__ and __git_version__ information into saved meta graphs. Change: 139471934
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a2f0f07f59b318d255ba86c8ab3486466dece5a8 |
|
04-Nov-2016 |
Zongheng Yang <zongheng@google.com> |
Remove a premature checkpoint existence test in Saver.restore(). The correct way to perform such a check is inside the Op kernel. Change: 138212174
/external/tensorflow/tensorflow/python/training/saver_test.py
|
194e39853943e5d87ce01ad8ce98def8fe0785e1 |
|
04-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Switch callers of tf.pack and tf.unpack to call tf.stack and tf.unstack instead. Change: 138139542
/external/tensorflow/tensorflow/python/training/saver_test.py
|
cbd3cacfb73bbea912b9d01c2540187684f751a7 |
|
03-Nov-2016 |
Illia Polosukhin <ipolosukhin@google.com> |
Replace usages initialize_all_variables -> global_variables_initializer Change: 138128703
/external/tensorflow/tensorflow/python/training/saver_test.py
|
5399bfaebfb4666b01ee6afc76dbf0455731bddd |
|
01-Nov-2016 |
Sherry Moore <sherrym@google.com> |
Added clear_devices option for export_scoped_meta_graph(). Change: 137767898
/external/tensorflow/tensorflow/python/training/saver_test.py
|
cc1acc3e398c65612ecfe865d6a0607f2f3de7ce |
|
24-Oct-2016 |
Sherry Moore <sherrym@google.com> |
Added clear_devices usage examples to saver_test and meta_graph_test. Added documentation for using clear_devices and to reset the default graph if one is to export and import meta_graph in the same default graph. Change: 137052401
/external/tensorflow/tensorflow/python/training/saver_test.py
|
dd845147f16679033f5efe3cbde7c417b3c71a8e |
|
21-Oct-2016 |
Dan Mané <danmane@google.com> |
Modify tensorflow/python/training to use the new summary ops. Very simple migration replacing tf.scalar_summary with tf.summary.scalar, etc. It is almost entirely a namespace shift. Also, the new scalar_summary does not support array of inputs, which was only really used in test code. Change: 136782705
/external/tensorflow/tensorflow/python/training/saver_test.py
|
6e8bef5a95a8bb826e4329e7ab586f79bb290fa1 |
|
21-Oct-2016 |
Sherry Moore <sherrym@google.com> |
Added ability to export, import and copy a scoped meta graph. Added functions to save a scoped meta graph and restore it into a different scope. Change: 136766406
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a21a900f2d014561914aa77268368e71a5f25a9d |
|
20-Oct-2016 |
Zongheng Yang <zongheng@google.com> |
TF Checkpoint V2: adapt saver_test to cover V2. This change adapts some unit tests, that previously would fail when V2 becomes the default. They now pass with both V1 or V2 being the default. This is in preparation for the upcoming switch to V2 (~end-of-month). Change: 136663731
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0c09f806b32bc0d3d9d6929033de70ac602a85c4 |
|
15-Oct-2016 |
Rohan Jain <rohanj@google.com> |
Fixing SaverUtilsTest to make sure that the temp dirs are created and deleted cleanly after each run. This test fails when run repeatedly because it checks for non-existence of a file which is actually created in a previous run. Change: 136231860
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8b203064f8e1cd1035d31f1bd2f60fa3461d8b94 |
|
14-Oct-2016 |
Zongheng Yang <zongheng@google.com> |
Introduce tf.train.{checkpoint_exists,get_checkpoint_mtimes}(). These two helpers are agnostic to the ckpt format -- it handles the naming difference between V1 and V2. Change: 136181681
/external/tensorflow/tensorflow/python/training/saver_test.py
|
d78fded517f6b01f3338e9216be66ad7e94ca8a4 |
|
08-Oct-2016 |
Zongheng Yang <zongheng@google.com> |
tf.train.Saver: add option to pad the step numbers in ckpt filename. Controlled by the constructor arg, "pad_step_number". Off by default. Change: 135529994
/external/tensorflow/tensorflow/python/training/saver_test.py
|
60051ecd911dadae8afd088ccbcb2024ccfe0373 |
|
02-Oct-2016 |
Sherry Moore <sherrym@google.com> |
Allow graph to be passed in to write_graph. Change: 134894910
/external/tensorflow/tensorflow/python/training/saver_test.py
|
ecdf0b202a2bfcff7985e62da727397bd8c67a91 |
|
28-Sep-2016 |
Zongheng Yang <zongheng@google.com> |
TF Checkpoint V2: make CheckpointReader work with the V2 format. If the same checkpoint prefix identifies both a V1 checkpoint and a V2 checkpoint on disk, the V2 version takes priority -- which matches the same behavior as the RestoreV2 op. Typical usage: $ bazel run tensorflow/python/tools:inspect_checkpoint -- --file_name=<V2 ckpt prefix> Other changes: add DebugString() and Contains() to BundleReader. Change: 134543092
/external/tensorflow/tensorflow/python/training/saver_test.py
|
1000807ffd1ce63ac2dd296b1990deb87a41d3d5 |
|
26-Sep-2016 |
Zongheng Yang <zongheng@google.com> |
TF Checkpoint V2: switch tf.train.Saver's restore op to RestoreV2. This prepares us for the upcoming V2 format: this change makes Saver backward-compatible with the V1 format, since the RestoreV2 op can read both V1 and V2 checkpoints. *IF THIS BREAKS YOU*: Likely you just need to replace some whitelist / golden file to include "RestoreV2" instead of "RestoreSlices". Change: 134306307
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a3c34f649d6f3d6c188cc59fe884facf4aad117e |
|
22-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Cleanup Change: 133963101
/external/tensorflow/tensorflow/python/training/saver_test.py
|
6333762303f2b858372ca779967d9fc1aa948b86 |
|
21-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
More detailed error message when saver files are not found. Minor fixes in related code/comments. Added test case to cover restoring from invalid paths. Change: 133819618
/external/tensorflow/tensorflow/python/training/saver_test.py
|
e91ea774cfce323ef5812274f92fa2f13d2f3637 |
|
13-Sep-2016 |
Sherry Moore <sherrym@google.com> |
Added to_proto and from_proto for CondContext and WhileContext. Change: 133013495
/external/tensorflow/tensorflow/python/training/saver_test.py
|
45e6676b1acb99a19052459a23ffb5437742d8e9 |
|
12-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add pylint indentation check to sanity and fix existing indentation Change: 132840696
/external/tensorflow/tensorflow/python/training/saver_test.py
|
1d2aa9451d920ad4bc8ad1ac86c06863b590e81f |
|
09-Sep-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Removes a few limitation of Defun: * Defun can appear anyway. Previously, it must be under a with tf.Graph() scope; Now, it can be used to decorate module level functions. * Defun constructs the function definition lazily so that if a Defun is not called, it cost little; * Defun can be used within another Defun. The inner ones are lifted to the top-level. Change: 132620258
/external/tensorflow/tensorflow/python/training/saver_test.py
|
51b23d50e3a3d021d54094bd25f22b9dde434202 |
|
25-Aug-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Set default '' for master arg, do not create a saver when the graph has no saveable variables. Also delete extraneous build(), as the saver does that in its constructor. Change: 131299934
/external/tensorflow/tensorflow/python/training/saver_test.py
|
b1413783a47bbf94dfbdd6e31d0e4a765a966d41 |
|
19-Aug-2016 |
Benoit Steiner <bsteiner@google.com> |
Merge changes from github. Change: 130695673
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0bfd671e7e310c17e12af23f8395c72db46292f9 |
|
10-Aug-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Change Saver to support checkpointing ops other than Variables. Operation save and restore is encapsulated in a SaveableObject class, which ops can implement to allow checkpointing. Add this support to MutableHashtable. Add an op for hash table that atomically clears the table and imports the data. Add a test for reshaping to saver_test.py Change: 129900685
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9996bdf4b8b325eac5013fa906f8c4294ea09ee6 |
|
10-Aug-2016 |
Mustafa Ispir <ispir@google.com> |
Added defered build option to the Saver to catch all variables just before finalizing the graph. Renamed _supervised_train => _monitored_train Change: 129823586
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f6bf341668bad321a351b62ecafe50bbbdadec2a |
|
03-Aug-2016 |
Zongheng Yang <zongheng@google.com> |
Typo fix in saver_test. Change: 129161698
/external/tensorflow/tensorflow/python/training/saver_test.py
|
323b57de58194202ba619a1301762f1e39189a6a |
|
29-Jul-2016 |
Derek Murray <mrry@google.com> |
Split TSAN-incompatible saver unit test into a separate target. Change: 128770169
/external/tensorflow/tensorflow/python/training/saver_test.py
|
c99a28959ca579e7464046247eaf6cd8d7e15206 |
|
18-Jul-2016 |
Zongheng Yang <zongheng@google.com> |
Fix failure to detect name collisions in Saver. This triggers when save_path is a relative path pointing to a file named "checkpoint", but does not trigger when it is an absolute path. Change: 127727572
/external/tensorflow/tensorflow/python/training/saver_test.py
|
10ea197604b6760652773d2525f850bfd3238061 |
|
13-Jul-2016 |
Zongheng Yang <zongheng@google.com> |
Partitioned vars: test for "save in N slices, restore in M slices" scenario. Where N != M. Previously, the Python tests were missing, so this functionality may not have been as widely exposed to users as we'd like. Change: 127276419
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8d594c8ec230f2e307aa7075284d3af4af10bd75 |
|
12-Jul-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Fix a bug when loading checkpoints that use relative path. Also add additional checks that can be triggered when users manually edit the checkpoint file and can help them debug potential mistakes they may have made. Change: 127138923
/external/tensorflow/tensorflow/python/training/saver_test.py
|
533d891cf64f1ad4b8470dd31141b1b1f642bae4 |
|
30-Jun-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Merge changes from github. Change: 126335170
/external/tensorflow/tensorflow/python/training/saver_test.py
|
40a85bcf69bf250736aeea70cf0c18e0fcd72daa |
|
28-Jun-2016 |
Gunhan Gulsoy <gunan@google.com> |
Make sure tests check if the machine has a GPU, rather than just checking if CUDA is linked in. Also fix a typo in tf.test Change: 126104437
/external/tensorflow/tensorflow/python/training/saver_test.py
|
c35042fd5332c9fdf205d9589756bcefcd3f30dc |
|
17-Jun-2016 |
Derek Murray <mrry@google.com> |
Make large-variable saving tests shorter to prevent timeouts. Change: 125132823
/external/tensorflow/tensorflow/python/training/saver_test.py
|
cc40cd3b0a8b83f5ee071b7ee32c17b56815a89c |
|
15-Jun-2016 |
Derek Murray <mrry@google.com> |
Add a size check before attempting to serialize a variable. This prevents the TensorSliceWriter from attempting to serialize variables that are larger than 2GB. It prevents potential memory corruption and segmentation faults. Fixes #2447. Change: 124921899
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0cf9ed3a719c0782695154d5a0bca260001cec15 |
|
02-Jun-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Update copyright for 3p/tf/python. Change: 123900456
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f48635800cb6991cbf82fc094c52a152dd90dd65 |
|
20-May-2016 |
Sherry Moore <sherrym@google.com> |
Added basic filesystem support in Saver to support gcs. Change: 122788931
/external/tensorflow/tensorflow/python/training/saver_test.py
|
94d27b6852b3e331fd9d64a0533f0fc27af05bfd |
|
06-May-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Next pass at converting all return statuses to errors.OpError exceptions in Python. - Add a tf_status util file to convert between Status and TF_Status. - Use TF_Status for the swigged APIs in session, checkpoint_reader, and server_lib. - Converts all these tf_status to exceptions in Python with the new context handler. - Remove the old StatusNotOK exception. Change: 121644982
/external/tensorflow/tensorflow/python/training/saver_test.py
|
46538d2b1fd63a2433633b860823535e79861c96 |
|
05-May-2016 |
Sherry Moore <sherrym@google.com> |
Set "add_shapes" to "True" when exporting graph_def in export_meta_graph. Fixes github issue 2166. Change: 121600220
/external/tensorflow/tensorflow/python/training/saver_test.py
|
95dc5bd1ee4ad415eee2b1f3601dc0baed379aa8 |
|
26-Apr-2016 |
Sherry Moore <sherrym@google.com> |
Added Howto for MetaGraph. Change: 120760402
/external/tensorflow/tensorflow/python/training/saver_test.py
|
747a3172dc931b5d2fa42c458870267126be51ef |
|
08-Apr-2016 |
Sherry Moore <sherrym@google.com> |
Added write_meta_graph flag to saver.save() to allow skipping writing meta_graph. Change: 119377154
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f3554041edfbadcde875abd3c762f23ff418f15d |
|
06-Apr-2016 |
Sherry Moore <sherrym@google.com> |
Added dtype information in DebugString(). Change: 119196706
/external/tensorflow/tensorflow/python/training/saver_test.py
|
286c5a48212d9eee522e026302524537e0e1cda8 |
|
01-Apr-2016 |
Sherry Moore <sherrym@google.com> |
Use all PEP8 names for CheckpointReader. The new APIs are debug_string() : returns debug string has_tensor(tensor_name): returns bool get_tensor(tensor_name) : returns numpy array for tensor get_variable_to_shape_map() : returns name to shape map Change: 118806412
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8baea485d6c3fafff8fd2d14cb0367318efcbd14 |
|
01-Apr-2016 |
Sherry Moore <sherrym@google.com> |
Added GetTensor() to return tensor content for a single variable from checkpoint file. To use: reader = tf.train.NewPyCheckpointReader(your-checkpoint-path) print(reader.GetTensor(your-tensor-name)) Change: 118792812
/external/tensorflow/tensorflow/python/training/saver_test.py
|
2966fcdae0cf7d75f4eca027f3003e355089b0ed |
|
23-Mar-2016 |
Sherry Moore <sherrym@google.com> |
Added Python APIs for inspecting checkpoint content. DebugString() HasTensor() GetVariableToShapeMap() Example use can be found in PywrapTensorSliceReaderTest in third_party/tensorflow/python/training/saver_test.py For example, To use (in Python): try: reader = tf.train.NewCheckpointReader(your_checkpoint_file) print(reader.DebugString()) except pywrap_tensorflow.StatusNotOK as e: print(str(e)) Change: 117951901
/external/tensorflow/tensorflow/python/training/saver_test.py
|
8ae38cb7d80ee4155c54ae3b704063def227b102 |
|
16-Mar-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Allows for importing of graphs with no variable nodes. Change: 117365836
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9d8ea873db7246656c21f5860b53e34f14dd2841 |
|
02-Mar-2016 |
Geoffrey Irving <geoffreyi@google.com> |
Expose tf.contrib.util.stripped_op_list_for_graph C++ and Python use two different op registries, and in rare cases they can actually be different. Thus, we need both functions available. Also fix both Python and C++ to handle arbitrarily nested functions. Change: 116142089
/external/tensorflow/tensorflow/python/training/saver_test.py
|
c1d385d01924cf02e81adabbdec575c100b14f59 |
|
01-Mar-2016 |
Geoffrey Irving <geoffreyi@google.com> |
Expose tf.contrib.util.stripped_op_list_for_graph C++ and Python use two different op registries, and in rare cases they can actually be different. Thus, we need both functions available. Also fix both Python and C++ to handle arbitrarily nested functions. Change: 115918836
/external/tensorflow/tensorflow/python/training/saver_test.py
|
d4b17f62fa09a10869d0f5d70e5afde833b23f0f |
|
01-Mar-2016 |
Geoffrey Irving <geoffreyi@google.com> |
Expose tf.contrib.util.stripped_op_list_for_graph C++ and Python use two different op registries, and in rare cases they can actually be different. Thus, we need both functions available. Also fix both Python and C++ to handle arbitrarily nested functions. Change: 115915068
/external/tensorflow/tensorflow/python/training/saver_test.py
|
4ecd2a70dd750b20a61033fe08301745685bf288 |
|
25-Feb-2016 |
Sherry Moore <sherrym@google.com> |
Added unit test for max_to_keep being None. Change: 115516426
/external/tensorflow/tensorflow/python/training/saver_test.py
|
cc53ecf5a91ae796cef823d5b9e5982a7bcc597e |
|
19-Feb-2016 |
Sherry Moore <sherrym@google.com> |
Added _MetaGraphFilename() function to return the name of meta_graph file. Remove the correct file in the case of sharded checkpoints. Change: 115015294
/external/tensorflow/tensorflow/python/training/saver_test.py
|
fe056f0b5e52db86766761f5e6446a89c1aa3938 |
|
17-Feb-2016 |
Vijay Vasudevan <vrv@google.com> |
Merge changes from github. Change: 114882676
/external/tensorflow/tensorflow/python/training/saver_test.py
|
7760ce56fc3ab4ab8cdc408e29d8ad8b539c417e |
|
11-Feb-2016 |
Josh Levenberg <josh11b@tensorflow.org> |
Get rid of some import cruft. Change: 114374558
/external/tensorflow/tensorflow/python/training/saver_test.py
|
18297126c0bc13c7045841f5a7a99f3da68176f4 |
|
10-Feb-2016 |
Geoffrey Irving <geoffreyi@google.com> |
Fix tf.test for PEP-8 and document tf.test now has appropriate snake case function names (get_temp_dir and is_built_with_cuda) and has normal toplevel module documentation. Also fix a bug in make_all. Change: 114351269
/external/tensorflow/tensorflow/python/training/saver_test.py
|
a4142f1289fea568b9fe8c0080bf4cdffa5a03ad |
|
10-Feb-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Includes ops used by user-defined functions in _stripped_op_list_for_graph. Change: 114294037
/external/tensorflow/tensorflow/python/training/saver_test.py
|
e52cecf69fa0efa782609a6d8f9494c08ae3aa7b |
|
09-Feb-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Added the use of google.protobuf.Any in meta_graph.proto to allow packing any any user protos. Change: 114256308
/external/tensorflow/tensorflow/python/training/saver_test.py
|
cc304733318f9635d37e72abc1c4d75b15a48d59 |
|
09-Feb-2016 |
Josh Levenberg <josh11b@tensorflow.org> |
Add the Ops for the producer to MetaGraphDef. Change: 114175259
/external/tensorflow/tensorflow/python/training/saver_test.py
|
0269c3690b402c63100869cbb7f8a8ea190e8c0a |
|
06-Feb-2016 |
A. Unique TensorFlower <nobody@tensorflow.org> |
Added export_meta_graph() and import_meta_graph() for serializing/de-serializing the graph and other Python objects necessary restarting training, running eval, or running inference into a MetaGraphDef protocol buffer. MetaGraphDef contains the following: - MetaInfoDef: For storing version and other meta data associated with the meta graph. - GraphDef: The Graph. - SaverDef: The Saver. - CollectionDef * Int64List * FloatList * BytesList * NodeList * AnyList These are evolving APIs and subject to change. Change: 114026857
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9b70316263eb74476ab96b7c0f300c4d90223425 |
|
25-Jan-2016 |
Vijay Vasudevan <vrv@google.com> |
Running our linter on a lot of files. Change: 112920860
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f9d3e9d03c69bfac77a2fe1ad80f7c5aa517e0f0 |
|
06-Dec-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: upstream latest changes to git. Change 109537918 TensorFlow pip setup: wheel >= 0.26 for python3 pip install Change 109505848 Fix distortion default value to 1.0 in fixed_unigram_candidate_sampler. This means we default to the actual provided unigram distribution, instead of to the uniform (as it is currently). Change 109470494 Bugfix in gradients calculation when the ys rely on each other. Change 109467619 Fix CIFAR-10 model to train on all the training data instead of just 80% of it. Fixes #396. Change 109467557 Replaced checkpoint file with binary GraphDef. Change 109467433 Updates to C++ tutorial section. Change 109465269 TensorFlow: update documentation for tutorials to not assume use of bazel (when possible). Change 109462916 A tutorial for image recognition to coincide with the release of the latest Inception image classification model. Change 109462342 Clear control dependencies in variable_scope.get_variable() when creating ops for the initializer. Add tests of various error conditions. Change 109461981 Various performance improvements in low-level node execution code paths. Speeds up ptb_word_lm on my desktop with a Titan X from 3638 words per second to 3751 words per second (3.1% speedup). Changes include: o Avoided many strcmp operations per node execution and extra touches of cache lines in executor.cc, by making all the various IsMerge, IsSwitch, IsSend, etc. operations instead be based on an internal enum value that is pre-computed at Node construction time, rather than doing string comparisons against node->type_string(). We were doing about 6 such comparisons per executed node. o Removed mutex_lock in executor.cc in ExecutorState::Process. The lock was not needed and the comment about the iterations array being potentially resized is not true (the iterations arrays are created with a fixed size). Checked with yuanbyu to confirm this. o Added new two-argument port::Tracing::ScopedAnnotation constructor that takes two StringPiece arguments, and only concatenates them lazily if tracing is enabled. Also changed the code in platform/tracing.{h,cc} so that the ScopedAnnotation constructor and the TraceMe constructor can be inlined. o In BaseGPUDevice::Compute, used the two-argument ScopedAnnotation constructor to avoid doing StrCat(opkernel->name(), ":", op_kernel->type_string()) on every node execution on a GPU. o Introduced a new TensorReference class that just holds a reference to an underlying TensorBuffer, and requires an explicit Unref(). o Changed the EventMgr interface to take a vector of TensorReference objects for EventMgr::ThenDeleteTensors, rather than a vector of Tensor objects. o Used TensorReference in a few places in gpu_util.cc o Minor: switched to using InlinedVectors in a few places to get better cache locality. Change 109456692 Updated the label_image example to use the latest Inception model Change 109456545 Provides classify_image which performs image recognition on a 1000 object label set. $ ./classify_image giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca (score = 0.88493) indri, indris, Indri indri, Indri brevicaudatus (score = 0.00878) lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens (score = 0.00317) custard apple (score = 0.00149) earthstar (score = 0.00127) Change 109455002 TensorFlow: make the helper libraries for various models available in the pip package so that when users type: python translate.py ... the absolute import works. This change is supposed to help make our tutorials run without the *need* to use bazel. Change 109450041 TensorFlow: remove cifar and convolutional binary copies from pip install. Adds embedding and some other models to the list. Change 109448520 Move the description of a failing invariant from a comment into the dcheck-fail message text. Change 109447577 TensorBoard has release tagging (tensorboard/TAG) Also track TensorBoard changes (tensorboard/CHANGES) Change 109444161 Added ParseSingleSequenceExample + python wrappers + unit tests. Change 109440864 Update all the TensorFlow Dockerfiles, and simplify GPU containers. This change updates all four of our Dockerfiles to match the targets discussed in https://github.com/tensorflow/tensorflow/issues/149. The most notable change here is moving the GPU images to use the NVidia containers which include cudnn and other build-time dependencies, dramatically simplifying both the build and run steps. A description of which tags exist and get pushed where will be in a follow-up. Change 109432591 Some pylint and pydoc changes in saver. Change 109430127 Remove unused hydrogen components Change 109419354 The RNN api, although moved into python/ops/, remains undocumented. It may still change at any time. Base CL: 109538006
/external/tensorflow/tensorflow/python/training/saver_test.py
|
854f49bd43588c062b046384f239f64a3d819702 |
|
25-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Upstream changes to git Changes: - Updates to docs - Several changes for Python 3 compatibility - Added license headers Base CL: 108710566
/external/tensorflow/tensorflow/python/training/saver_test.py
|
9c3043ff3bf31a6a81810b4ce9e87ef936f1f529 |
|
20-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Improve performance of Alexnet Changes: * error message that refers to removed `DefaultSession` method. * -Wnull-conversion warnings * the "_start_time" attr for recvs when the flag "--brain_enable_scheduling_for_recvs" is set. * typo in tutorial data download progress message. * a typo ("however their installing"=>"however installing"). * typo, rename "TensorFlow Mechanics" to "How To" to be consistent with the website. * a typo ("subtact"=>"subtract"). * protobuf examples in comments in tensorflow::Example.proto. * formula formatting in MNIST beginner tutorial * negative fraction-of-queue-full stats * protobuf inclusion path so that Android demo will build under Blaze. * small typo (moderatly > moderately) * Session.run() to check that tensor arguments come from the session's graph. * another six import * seq2seq typo in bazel command Base CL: 108349164
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f7918e1dcd5b0c1f8114f488fc35a63a81e94535 |
|
19-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: Removal of large assets and small other fixes. Changes: - Remove all large assets from the repoistory, incuding the other 50MiB model protobuf and a lot of images in our g3doc directory. We will maintain these assets externally for now. g3doc images may be broken for a little bit, but the website will be fine, which is the important resource. By @vrv and @petewarden. Updates READMES to reflect the external model resources. - Fix to saver's latest_checkpoint function by Zhifeng - Made protos visibility public by @vrv - Updates to docs by @mrry, Andy - Embed tensorboard resource for summary icon by Daniel - More updates to backwars compat by @josh11b Base CL: 108194981
/external/tensorflow/tensorflow/python/training/saver_test.py
|
011e9baccd343eb943d25014c4e8aec53eac396b |
|
13-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: a few small updates. Changes: - Fix softmax formula in word2vec to remove an extra exp() by @gouwsmeister - Python3 fixes to remove basestring / support for unicode by @mrry - Remove some comments by Josh - Specify exact versions of bower dependencies for TensorBoard by @danmane. Base CL: 107742361
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f2102f4e2c1c87f1d1bf9ab856a2849c54478760 |
|
12-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: upstream changes from the afternoon. Changes: - futurize --stage2 changes for Python 3 compatibility by @girving. - Small updates to documentation by @vrv, schuster and others - Account for failure of std::thread::hardware_concurrency by @ebrevdo. - More changes for backwards-compatibility tests by Josh - Updates to python op doc generation by Josh - Added support for using the best-fit allocator via ConfigProto by @vrv. - Rename LocalSession to DirectSession, since local was a bad name for it. - Enable tf.nn.moments() to work with tensors of unknown shape by @mrry. GITHUB_ISSUE: 139 - Changes for Android build by Andrew. Base CL: 107645181
/external/tensorflow/tensorflow/python/training/saver_test.py
|
61d3a958d6d83cb6037490d933b47621cc4009cc |
|
09-Nov-2015 |
Vijay Vasudevan <vrv@google.com> |
TensorFlow: Initial steps towards python3 support, some documentation bug fixes -- reindents to 2 for some of the files to match our internal requirements. Thanks to Martin Andrews for the basic_usage.md suggested fix via Gerrit. Base CL: 107394029
/external/tensorflow/tensorflow/python/training/saver_test.py
|
f41959ccb2d9d4c722fe8fc3351401d53bcf4900 |
|
07-Nov-2015 |
Manjunath Kudlur <keveman@gmail.com> |
TensorFlow: Initial commit of TensorFlow library. TensorFlow is an open source software library for numerical computation using data flow graphs. Base CL: 107276108
/external/tensorflow/tensorflow/python/training/saver_test.py
|