hash
stringlengths
40
40
msg
stringlengths
1
131k
author
stringlengths
1
33
email
stringlengths
0
57
date
int64
1,447B
1,698B
a466e6491001464843a5ce69aeebd1a940786adc
Update to absoulte imports in skflow
Illia Polosukhin
ilblackdragon@gmail.com
1,458,025,253,000
2d580ccb2b13bedbaed0cc38fbf970721607584f
Adding more from __future__ imports and fix aboslutes in monitors
Illia Polosukhin
ilblackdragon@gmail.com
1,458,027,921,000
50d745d2a3058da2cb8d803bb05eff2c52b6997d
add backoff for cPickle to pickle for py3
Illia Polosukhin
ilblackdragon@gmail.com
1,458,028,545,000
f7a0b752131dc145191521730779e096864efc1d
Remove unused file
Martin Wicke
martin.wicke@gmail.com
1,458,028,771,000
27af299f5e0da600a4431ed0f5ecc59ae793e31c
typos fix and ign temp files in gitignore
rasbt
mail@sebastianraschka.com
1,458,011,356,000
13ab7ff23f645157db978122baf5f4ca0af25ca8
Add complex128 dtype to framework
Igor Babuschkin
igor@babuschk.in
1,457,395,293,000
37bf6ce4864f1ce4344f68556d37f6105faf989b
Fix error in py3 for writing bytes in to file with 'w'
Illia Polosukhin
ilblackdragon@gmail.com
1,458,059,625,000
fa4346762ebf9a8fdc6f24a1f580c058f4b45f84
Fix for reading byte for py3
Illia Polosukhin
ilblackdragon@gmail.com
1,458,060,073,000
f92e338b178f75b0e7fc79aedc00680a49a982ac
Merge pull request #1425 from ibab/implement-complex128 Add complex128 dtype
Geoffrey Irving
irving@naml.us
1,458,064,367,000
a6e174fb09b1841110d00c4321bb974917bc495a
When selective registration is used, only register gradient functions if there is a SymbolicGradient op present. Fine-grained registration is not needed presently as the production inference models will not have gradients. Change: 117178512
A. Unique TensorFlower
nobody@tensorflow.org
1,457,994,510,000
fc088fa9608bbde4074e5aa3768b36270cb00753
Fixes Android build by ifdef'ing out the use of RTLD_NOLOAD. Change: 117181098
Derek Murray
mrry@google.com
1,457,996,070,000
e96bc164c772456e6298bc5ea0a9005c296ebe59
Whitespace bugfix in tf_py_tests Change: 117185329
Eugene Brevdo
ebrevdo@gmail.com
1,457,998,383,000
ac4b54e2171e27937948108cd42cedc86d325710
Make tf_custom_op_library depend on cudart_static directly instead of build_config:cuda. This makes it possible to expand build_config:cuda to include other cuda dependencies not compatible with tf_custom_op_library. Change: 117186282
Manjunath Kudlur
keveman@gmail.com
1,457,998,900,000
ca9bb0ccb188865cb34b0dc4b2a017f84ecf3d5c
Add abort op to end a process when needed. Change: 117191014
Jianmin Chen
goog.jmchen@gmail.com
1,458,001,367,000
9795521f98b48bc9c894317d0bea70100dfbe36d
Ensure that keys, in HashTable::Find, are validated in the face of asynchronous updates to the keys tensor. Change: 117194227
Yutaka Leon
yutaka.leon@gmail.com
1,458,003,533,000
bc964ae42fb1f661bc5452010a185434d2dbf75d
Fix stale uses of the use_locking attr that could allow unlocked access to the ref tensor itself. With this change, all locks are grabbed, and duplicates are explicitly skipped. Locks are acquired in address order to provide a lock discipline for avoiding deadlock. (Note that this is not guaranteed effective unless ...
David G. Andersen
dga@google.com
1,458,004,731,000
c7f80ca7fd13e131f9be55c182033fad269ec4ad
Change tf_gpu_kernel_library() and tf_cuda_library() rules to not depend on private: tensorflow/core/platform/google/build_config:cuda_runtime_extra Change: 117197032
Josh Levenberg
josh11b@tensorflow.org
1,458,005,529,000
f9905bf233f452bf405da6278e55e30600d065f2
Reverting use_locking change temporarily to fix test failure. Change: 117208315
David G. Andersen
dga@google.com
1,458,019,618,000
460ef5f4710bdf5e1a9b1b4471dd0605d7c1882e
Make the requirements for docs callouts clearer. Change: 117209040
Martin Wicke
wicke@google.com
1,458,020,709,000
af4e6fc72586239d73402564042d5ad4aa640d22
Fix //tensorflow/python:server_lib_test in Python 3. Adds a typemap for the `const char* target` argument to `TF_SetTarget()` that accepts any object that may be converted to a null-terminated string. Change: 117241091
Derek Murray
mrry@google.com
1,458,055,664,000
9c7be1cf1c4f8cead83e8137a9d0718b657995af
Fix python 3 vs 2 issue regarding bytes != str Was able to reproduce the issue in python3. The problem comes down to 'str' and 'bytes' being equivalent types in python2, but in python3 they are not, since python3 encodes all strings as unicode. Because of this, they need to explicitly be converted to bytes using an en...
Dan Smilkov
dsmilkov@gmail.com
1,458,058,034,000
27fb538613e25ec00fba565d8603d1a841de3341
Handle the case when input size is zero. Change: 117245363
Manjunath Kudlur
keveman@gmail.com
1,458,058,866,000
9b7cb19d57865eabca7e950fc3b0598d81ec5e63
Add a function to compute the sum of multiple variable regularizers. This makes it possible to apply regularizers like the Elastic Net, which incorporates separate L1 and L2 regularization components. Change: 117246793
A. Unique TensorFlower
nobody@tensorflow.org
1,458,059,871,000
ac54660e70b1a49acc813858b56d18fa65b168b9
Ensuring test coverage for the use_locking=True path Change: 117250550
David G. Andersen
dga@google.com
1,458,062,059,000
c9c341e4833612392ddd4fca3fd4343afaed47a4
Ensure that keys and values, in HashTable::Insert, are validated in the face of asynchronous updates to the keys and values tensors. Change: 117252477
Yutaka Leon
yutaka.leon@gmail.com
1,458,063,074,000
b10c65c2a00b9ce7972ab4300f49e83d0f849fd4
Remove use of conitional pass-through from dynamic_rnn in favor of just calling select() at each step. conditionals add a bunch of extra ops that slow things down; and generally the size of the input tensor into dynamic_rnn matches max_sequence_length so they provide no benefit. Before change, benchmarks: Calculatio...
Eugene Brevdo
ebrevdo@gmail.com
1,458,064,349,000
725e968a934419eec45b9acaf4bf8dc5f5f0574e
Remove race conditions from TensorShapeUtils::MakeShape Also move some functions from header to C++. Change: 117255894
Geoffrey Irving
geoffreyi@google.com
1,458,064,951,000
b27022dd3ccb0b415cef3505ee0ee65f25b2ffcb
Fix GetOpList and GetPythonWrappers SWIG wrappers for Python 3. - Return the result from GetOpList as uninterpreted bytes object. - Write a input typemap for GetPythonWrappers to receive python 'bytes' object and convert to const char* pointer and length. Change: 117258253
Manjunath Kudlur
keveman@gmail.com
1,458,066,011,000
ab3119e565e762828a5a818de8cdab61e5056fec
Fix bug concerning computation of gradients inside a CondContext. Change: 117260468
A. Unique TensorFlower
nobody@tensorflow.org
1,458,067,142,000
d2803a17d9c0432cbb232894cfb74979b19bbe46
Merge branch 'master' into skflow
Illia Polosukhin
ilblackdragon@gmail.com
1,458,071,068,000
9a8c5ad18c61cb0695d31e2ce969008c82999c7c
Prevent the feeding of tensors whose values are used to calculate shapes. This change prevents feeding a tensor if its constant value has been accessed. For example, the constant value is often used in shape functions (e.g. in `tf.reshape(data, indices)`, `indices` is a tensor, but it is often constant) in order to in...
Derek Murray
mrry@google.com
1,458,068,542,000
80159cde9ac0bba6e08d6694ece60e9ebe87235a
Add resize_nearest_neighbor_op.cc to extended Android op filegroup. Change: 117263085
Andrew Harp
andrew.harp@gmail.com
1,458,068,571,000
c33b697cc7c6b748386190a9184d688ea53e43a1
Removing images from documentation for now as per hosting requirements on tensorflow.org.
Illia Polosukhin
ilblackdragon@gmail.com
1,458,072,736,000
4424d1034c6097614b8d880049a6f5e5400350fb
Merge branch 'skflow' of github.com:ilblackdragon/tensorflow into skflow
Illia Polosukhin
ilblackdragon@gmail.com
1,458,072,744,000
8ebf5c2309d6f488db17c6e972cc6c0afc232e48
Merge pull request #1393 from panmari/nn_grad Added GPU implementation for resize nearest neighbor grad.
Martin Wicke
martin.wicke@gmail.com
1,458,072,930,000
0513a3b6d3c78e379e118b9f1f70b5320c06d388
rint op
Alan Wu
alan.c.wu@gmail.com
1,456,953,809,000
9215bc32f615af47a32958ef5c9142504632e91e
Merge pull request #1445 from ilblackdragon/skflow Adding skflow to tensorflow/contrib
Martin Wicke
martin.wicke@gmail.com
1,458,074,303,000
6f1acd3af3493de4cb7900629ce065d66c8c28bc
A few more tweaks to test_log.proto to reflect what's accessible from bazel build. Also some updates to run_and_gather_logs. Reduce the set of tests that run_and_gather_logs_test runs to speed up the tests. Change: 117271064
Eugene Brevdo
ebrevdo@gmail.com
1,458,073,324,000
9f6dd926d6b20bfd8edcd5589c7cf02ddb1804f8
Log the names of the operations that can't be colocated to help debugging. Change: 117273353
A. Unique TensorFlower
nobody@tensorflow.org
1,458,074,621,000
e42c2af56b054194cd72f46a717fb9a5506071a4
Fix inconsistency in size inference for RNN cells. Do input size inference based on inputs to LSTMCell, the same as for other cells. Leave the broken input_size @property in place as for the other cells. Change: 117276425
A. Unique TensorFlower
nobody@tensorflow.org
1,458,076,194,000
0d18da7bdfae192896e87adb8a69199f9b1386b5
Removing the check for duplicate ids. Change: 117280722
A. Unique TensorFlower
nobody@tensorflow.org
1,458,078,574,000
91e6443a06348c50b99848572352b039b3209242
Merge branch 'master' into translate_fix
Yuwen Yan
ybbaigo@gmail.com
1,458,084,042,000
82633d9b890dcb64e31a6b43b6da9ec81ffdbc04
Corrected link to example and added link to new blog post
terrytangyuan
terrytangyuan@Gmail.com
1,458,097,217,000
d0a3000c9232f8b1d51ab050d0e46d897a33955c
fixes #1522 close file object properly for read_data()
ronrest
ronny.coding@gmail.com
1,458,110,058,000
61afcdbff23340ca17c64c24cd77fd07115ae0a1
changed predict to use self.batch_size by default in skflow added an example using gridsearchcv
Michael Heilman
mheilman@civisanalytics.com
1,458,138,515,000
0a51ae3f2a8596beba6d85507099124d573b83f3
Typo?
Yuxin Wu
ppwwyyxxc@gmail.com
1,458,139,660,000
1d6d565a3d49adaa1891145bc0b1b59d33bcc5cb
Merge pull request #1521 from terrytangyuan/link [skflow] Corrected link to example and added link to new blog post
Vijay Vasudevan
vrv@google.com
1,458,159,191,000
9fec529b7fcce4546afe618a42ae5b7c14b00dc5
Removes gradient error tests, and makes small change to create ranges with math_ops (and to avoid Python 3 errors). Change: 117284043
A. Unique TensorFlower
nobody@tensorflow.org
1,458,080,403,000
477312ea105663fa5d6ba79e49dcceea7437c1de
Re-introducing previous fixes to locking, taking advantage of additional test support for the use_locking path. END_PUBLIC Automated g4 rollback of changelist 117208315. *** Reason for rollback *** Fixed bug introduced in original CL *** Original change description *** Automated g4 rollback of changelist 117196108...
David G. Andersen
dga@google.com
1,458,080,687,000
127d17a7bc4b25235b105e0313f1b7d57728bed1
Avoid race conditions in TensorShape and PartialTensorShape Also fix MakePartialShape to correctly set is_unknown_ = false. Change: 117285169
Geoffrey Irving
geoffreyi@google.com
1,458,081,063,000
be59d7a4eef8b914ebb4aa51124613f032391a3d
fix error message formatting in session_manager.py Change: 117285646
A. Unique TensorFlower
nobody@tensorflow.org
1,458,081,318,000
a0d21ec39cc3f18781d2d37798aa328e12f92844
Disable //third_party/tensorflow/core/distributed_runtime:master_test. This test is flaky. We need a more robust solution for starting multiple processes and connecting them; `PickUnusedPortOrDie()` is not robust to multiple concurrent tests on the same host. A possible approach would be to add a portserver, or add a ...
Derek Murray
mrry@google.com
1,458,086,345,000
3ae663ccc5d08976e0f547d5b2ece35067a6673e
Merge changes from github. Change: 117301677
Martin Wicke
wicke@google.com
1,458,093,213,000
e06a4a2e1dc0dc6f36f63f13bb048c8cbb2b2c9c
- Add a generic immutable hash table op that is initialized once and used to map key tensors to values. - Add a table initializer given the keys and values as tensors. Example use case: keys = tf.constant([0, 1], tf.int64) values = tf.constant(["hello", "world"]) default_value = "UNK" table = tf.contrib.loo...
Yutaka Leon
yutaka.leon@gmail.com
1,458,093,602,000
fdc6752cda33e8d5879e4db68093eca7d7395988
Add tf.contrib.lookup.string_to_index and tf.contrib.lookup.index_to_string to map strings to IDs and viceversa. Change: 117303981
Yutaka Leon
yutaka.leon@gmail.com
1,458,096,117,000
9a4878cc17d4039939e8df5a2a984cca8028baeb
Rollback of: "Merge changes from github." Change: 117304114
Vijay Vasudevan
vrv@google.com
1,458,096,327,000
0e73901fc1e643ddea8e1d0b3a1df018674c2074
Adds scalar_logistic_loss() and apply_regularization() to contrib Change: 117305077
Wei Ho
weiho4+github@gmail.com
1,458,097,583,000
f4ade48bc73257724e0069b3e2cdf41e96fcc32c
C++ gradients: split functions gradient code out into its own file. Change: 117346884
A. Unique TensorFlower
nobody@tensorflow.org
1,458,143,510,000
7be94c2f2987db1d45a0d90f56865b0b6a2c93e0
Fix another unused code warning. Change: 117347879
Josh Levenberg
josh11b@tensorflow.org
1,458,144,222,000
1fa0d1eaf5cbb598cbf89dddf8e5729ef22bc5ae
Eliminating the DoValidation/DoCompute split for training kernels and eliminating several other asynchrony-related validation bugs. (This turns out to also shrink the code a bit by eliminating a fair bit of redundant input accessor code between the validation and compute phases). Haven't benchmarked, but if anything, ...
David G. Andersen
dga@google.com
1,458,147,545,000
ca7eac9d8d4803a58ff5c915ed3498b4aefacbea
Extending tensorflow::Env with support for memmapped files Change: 117356696
A. Unique TensorFlower
nobody@tensorflow.org
1,458,149,478,000
ca0909cdfcf009eb418302774b1caf28c50ade35
Change dep list of tf_gpu_kernel_library rule. Change: 117361209
A. Unique TensorFlower
nobody@tensorflow.org
1,458,151,679,000
0d0d92ca52c3f1ce24edc29b60bf1e41c382d393
Fixing an int64->int32 conversion that may have made concatoffset not work with (ridiculously) large tensors. Also switching an existing test to use FastBoundsCheck to increase consistency with where we're taking the other ops. Change: 117364143
David G. Andersen
dga@google.com
1,458,153,144,000
8ae38cb7d80ee4155c54ae3b704063def227b102
Allows for importing of graphs with no variable nodes. Change: 117365836
A. Unique TensorFlower
nobody@tensorflow.org
1,458,154,038,000
f1bccd320a742c67245ded0cbe5eba971294d769
Another bugfix in run_and_gather_logs. Change: 117375431
Eugene Brevdo
ebrevdo@gmail.com
1,458,159,563,000
4c85a08666796faad743a47b63d350132d5c0c90
Rollforward of "Merge changes from github." Change: 117375570
Geoffrey Irving
geoffreyi@google.com
1,458,159,634,000
769aa524c4bdc99c61df468257c95c2f7daeb16e
Because TensorFlow knows the entire graph of your computations, it can automatically use the [backpropagation algorithm](http://colah.github.io/posts/2015-08-Backprop/) to efficiently determine how your variables affect the cost you ask it to minimize. Change: 117382393
A. Unique TensorFlower
nobody@tensorflow.org
1,458,163,129,000
db511114eafa28fd5d023f11279236dbf28f98ba
Simplify pylint disables. Change: 117384554
Josh Levenberg
josh11b@tensorflow.org
1,458,164,170,000
2ce60da7d2ece5d23b0843bce42123c7b5633c71
Forcing copy of input tensor, switching to use FastBoundsCheck Change: 117384840
David G. Andersen
dga@google.com
1,458,164,332,000
a3ef0fdaacd18fffd1e77bca47596c0b8ddc9775
Move Relu Op declarations into relu_op.h. Change: 117387358
A. Unique TensorFlower
nobody@tensorflow.org
1,458,165,636,000
d68382074d9ca483e044d1fa3c9851b96ad51482
Move Relu Op declarations into relu_op.h. Change: 117392494
A. Unique TensorFlower
nobody@tensorflow.org
1,458,168,312,000
6f41f6e3a40f3a782cbacabd00ee89d08bec0095
Fixing a test for posix environment for memmapped regions Change: 117394750
A. Unique TensorFlower
nobody@tensorflow.org
1,458,169,664,000
26fd3e4d900a940def1bc7c11dd65183dd0401fd
Merge pull request #1288 from chemelnucfin/rintop Rint Operator
Martin Wicke
martin.wicke@gmail.com
1,458,171,486,000
df2ea2c8fb6fc2a653843ec521adb52305f1679f
Merge commit for internal changes
Vijay Vasudevan
vrv@google.com
1,458,172,820,000
945a4b4cd12df0fdf3f0cc24b128bb4ce66ebf88
Revert "rint op" This reverts commit 0513a3b6d3c78e379e118b9f1f70b5320c06d388. This seems to break the GPU build with recent changes, see http://ci.tensorflow.org/job/tensorflow-master-gpu_pip/345/changes
Vijay Vasudevan
vrv@google.com
1,458,174,479,000
9be9a3dfdbc333134a5c417d97fae4629f14cd83
Fixes #skflow/148 - off by one error
Illia Polosukhin
ilblackdragon@gmail.com
1,458,080,769,000
d868f1e8089b2cbee7c7c7f6b385dda9a13c1c7e
Merge pull request #1520 from ilblackdragon/fix-148.
Derek Murray
mrry@google.com
1,458,225,810,000
4afc661767668f045689bff4b6aca42d2fa7b20f
fix typos
Fabrizio (Misto) Milo
mistobaan@gmail.com
1,458,180,105,000
6ad05ee6ff434b26238c9ba3558290fe7c4a4cf6
Merge pull request #1529 from Mistobaan/features/fix-mispells fix typos
Vijay Vasudevan
vrv@google.com
1,458,240,029,000
2b9c6e933098b5bff0f379e5f8c3951b588c7a44
initial logsoftmax kernel merged tests into a single test and added overflow test bazel run tensorflow/core/ops/compat:update_ops
Kashif Rasul
kashif.rasul@gmail.com
1,457,387,431,000
f71e2fd745cefcecb2bb662aa1ddb30f71928d1e
Merge branch 'master' into issue-443-logsoftmax
Dr. Kashif Rasul
kashif.rasul@gmail.com
1,458,252,031,000
a133a2d51670318aaab8ec7ec74534cecbddb03f
Merge branch 'master' into issue1522
Ronny
ronrest@users.noreply.github.com
1,458,252,231,000
d6c7954918f11beedd81d68df7a1626b149398da
Merge pull request #1527 from ppwwyyxx/patch-2 Fix mis-specified data format in documentation.
Vijay Vasudevan
vrv@google.com
1,458,253,163,000
ac7f85208bba1b2d12205e33f9b04d47fbdef726
Merge branch 'master' into issue1522
Ronny
ronrest@users.noreply.github.com
1,458,254,555,000
2085865f846240f57b58052b6b99eee0502ec00c
Fixing some python3 failures. Fixes much of tensorflow python tests and tensorboard server_test.
Vijay Vasudevan
vrv@google.com
1,458,260,352,000
4648c74ba0c0e6ef3f4313e964de1b41d3888188
Fix python3 compat issues for supervisor_test, remove use of internal functions, using public API instead.
Vijay Vasudevan
vrv@google.com
1,458,261,524,000
6bcf26f2a9bb0f6ab86f2cf5f84504ce6ab8a2d1
Disable device_lib_test due to swig-python3 incompatibility.
Vijay Vasudevan
vrv@google.com
1,458,261,952,000
23d4411ef5f0f772c072d2035083940f7605b6be
Fix lookup_ops_test for python3 compatibility
Vijay Vasudevan
vrv@google.com
1,458,262,890,000
524f0c587c07124eee6cbcbeff50688a041ad491
Regenerate ops.pbtxt from ops history.
Vijay Vasudevan
vrv@google.com
1,458,263,037,000
a8d97236a7b84ea00e5bb9aa5f2dcd2cab0e0899
Merge pull request #1537 from vrv/fix_python3 Fixing some python3 failures. Fixes much of tensorflow python
Vijay Vasudevan
vrv@google.com
1,458,265,022,000
4fd4ac6e7413b7e57a133a1ebc5eeccce4f6464b
Merge pull request #767 from Mistobaan/feature/explicit-forget-bias rnn_cell.py: Allow user to explicitly set forget bias in LSTMCell
Vijay Vasudevan
vrv@google.com
1,458,277,683,000
ff3ed0e58141a02c2be39cbdf15892ad7caeedba
Merge pull request #1496 from ybbaigo/translate_fix fix crash for empty dev_set bucket
Vijay Vasudevan
vrv@google.com
1,458,280,934,000
19ad6e56a39fb16719b93829a3546dfcfd479be9
Merge pull request #1424 from kashif/issue-443-logsoftmax Logsoftmax kernel
Vijay Vasudevan
vrv@google.com
1,458,280,963,000
75de3675286edff0e8c2f9994b13dbd4233cd7ae
Merge pull request #1455 from delftswa2016/Fixing-Broken-Link Fixed broken link in "Links"
Vijay Vasudevan
vrv@google.com
1,458,280,999,000
c05b4060f1e1261f3b11d0cc4cb0dcb8a57ede3b
Merge pull request #1417 from chemelnucfin/diaggrad gradient of diag operator
Vijay Vasudevan
vrv@google.com
1,458,281,164,000
6c28817b9bd443edea45c351ac8bab8f31e1ba54
Merge pull request #1523 from ronrest/issue1522 fixes #1522 close file object properly for read_data()
Vijay Vasudevan
vrv@google.com
1,458,281,436,000
bce5772169f575c99d29ccec35db69fe567c1168
ci_build: add cmake support
Jan Prach
jendap@google.com
1,457,920,028,000
57c078b5434ff63d49399e2fcb2f9a63e1ac8c37
install sklearn to debian
Jan Prach
jendap@google.com
1,458,286,530,000
1efc50a1da2472fc0349a81f344cd0463e68d505
fix virtualenv pip testing
Jan Prach
jendap@google.com
1,457,947,384,000
73def6d6a52058b634d6ded57d2fcfdf152e1881
Fix typos
Johnny Lim
izeye@naver.com
1,458,303,021,000