Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Clarify batch_norm documentation to highlight that the dimensions for… #15653

Closed
wants to merge 4,256 commits into from
Closed

Clarify batch_norm documentation to highlight that the dimensions for… #15653

Changes from all commits
Commits
Show all changes
4256 commits
Select commit Hold shift + click to select a range
11f39ed
Merge branch 'master' into sru
tjingrant Dec 14, 2017
94672d8
Merge pull request #15356 from dandelionmane/branch_178996911
gunan Dec 14, 2017
fbb334b
Merge pull request #15173 from JohanJu/master
caisq Dec 14, 2017
f158b7d
Revert "Add batch support for various image_ops (#14854)" (#15349)
jhseu Dec 14, 2017
c762adc
Benchmarks for flipping and random flipping (#15348)
jhseu Dec 14, 2017
b65bed5
Dockerfile.devel-gpu: optimize the size of the generated image (#15355)
flx42 Dec 14, 2017
162c39f
Merge pull request #14321 from yongtang/11052017-tf.Constant
martinwicke Dec 14, 2017
3dcf319
Changed ffmpeg verbosity semantics (#14582)
chrisdonahue Dec 14, 2017
7b71354
Add description for new PR workflow.
yifeif Dec 15, 2017
f454641
Merge pull request #14401 from myungjoo/fix/options
martinwicke Dec 15, 2017
a7291f3
Merge branch 'master' into fixed-typos
martinwicke Dec 15, 2017
40593a0
Merge pull request #14399 from PW486/fixed-typos
martinwicke Dec 15, 2017
752462b
Merge pull request #14389 from cthom055/master
martinwicke Dec 15, 2017
a933e72
address comments and explain constraint
tjingrant Dec 15, 2017
a4071e9
Instead of "option" use "set" to define non-bool cmake build args. (#…
gunan Dec 15, 2017
c9eda94
address pylint concern
tjingrant Dec 15, 2017
89f0ae5
fix wrong merge
tjingrant Dec 15, 2017
55cb118
Fix api usage in examples of gan
betterenvi Dec 15, 2017
c6fbdea
Fix typo of tf.abs docstring
qmick Dec 15, 2017
bdaffd1
- Bumping abseil-cpp to latest for FreeBSD compat
Dec 15, 2017
ab6c1f6
rebasing
Dec 15, 2017
d2cd2f5
rebasing
Dec 15, 2017
8073201
Support --config=monolithic in tf.sysconfig.get_link_flags() (#15139)
alsrgv Dec 15, 2017
9e07906
Merge pull request #14510 from facaiy/BUG/index_to_string_table_from_…
martinwicke Dec 15, 2017
3b759b9
Update CONTRIBUTING.md
yifeif Dec 15, 2017
bdb3438
Merge pull request #15377 from tensorflow/yifeif-patch-1
martinwicke Dec 15, 2017
834bd0b
Add back blank line
martinwicke Dec 15, 2017
36b0b24
Add `decode_libsvm` for libsvm format support
yongtang Nov 7, 2017
2d9572f
Add python wrapper for decode_libsvm
yongtang Nov 7, 2017
7e169c5
Add bazel BUILD and test file
yongtang Nov 7, 2017
a42d1a9
Add shape inference and upate tests
yongtang Nov 7, 2017
20e7e65
Update tests to use sparse tensor
yongtang Nov 15, 2017
e485d35
Update kernel to use sparse tensor
yongtang Nov 15, 2017
e7aeb21
Expand to 2+-D inputs
yongtang Nov 15, 2017
f05bd6b
Add num_features_ >= 1 constraint.
yongtang Nov 15, 2017
09adabe
Address review feedback
yongtang Nov 21, 2017
c1c3a20
Add different dtype support for label
yongtang Nov 21, 2017
f655c8d
Update tensorflow/contrib/cmake/tf_core_kernels.cmake and tensorflow/…
yongtang Nov 21, 2017
a055486
Update return values to `features, labels` to be consistent with othe…
yongtang Nov 29, 2017
2c051ae
Replace loop iteration with `chip` (#15289)
yongtang Dec 15, 2017
1f99137
GPU: Add Complex kernel for tf.exp() (#15182)
facaiy Dec 15, 2017
dd51440
[Go]: Make op wrapper generation more robust. (#15353)
asimshankar Dec 15, 2017
c0373a0
Bump the eigen dependency version. (#15405)
gunan Dec 15, 2017
d854f2e
Add compression support to RecordInput
yongtang Aug 17, 2017
7e934c0
Add test cases of compression_type for RecordInput.
yongtang Aug 17, 2017
b9e9d58
Load boundaries array into shared memory before hand for `bucketize` …
yongtang Dec 16, 2017
878ecb2
Update API with `tensorflow/core/api_def/update_api_def.sh`
yongtang Dec 16, 2017
0b886e3
change order of matrix multiplications to avoid large matrices
vnavkal Dec 16, 2017
3aad34a
remove TODO comment describing a task that seems infeasible
vnavkal Dec 16, 2017
67da46c
Add customized kernel implementation for clip_by_value (#13998)
yongtang Dec 16, 2017
aae1d45
Initial SRU Implementation (#13978)
tjingrant Dec 17, 2017
a21473f
Fix issues in doc `tf.Placeholder` should be `tf.placeholder` (#15330)
yongtang Dec 17, 2017
bd832b6
Add common error documentation
mikeknapp Dec 17, 2017
96e4775
Merge pull request #15333 from taehoonlee/fix_typos
caisq Dec 17, 2017
0b2d4b4
Merge pull request #15313 from rajendraarora16/patch-5
caisq Dec 17, 2017
7845836
Revert "Initial SRU Implementation (#13978)"
gunan Dec 18, 2017
7e5da07
Revert "Add customized kernel implementation for clip_by_value (#13998)"
gunan Dec 18, 2017
958d11f
Fix PEP8 (#15378)
taehoonlee Dec 18, 2017
901d9a7
Fix a bug: bfloat16 is unsigned on Windows (#15302)
snnn Dec 18, 2017
4f61c11
Add an is_external arg to tf_copts
snnn Dec 18, 2017
9c6a121
Return empty string when path resolver fails
Dec 18, 2017
0e43872
Fix lib_strings_str_util_test on Windows
snnn Dec 18, 2017
71545ce
Update math_ops.py
AndreiCostinescu Dec 18, 2017
bf56c8a
Update core.py
AndreiCostinescu Dec 18, 2017
c70b60b
Fix the CODEOWNERS file syntax (#15411)
nealwu Dec 18, 2017
e5efe26
Merge pull request #12369 from yongtang/12344-RecordInput-compression
martinwicke Dec 18, 2017
7acf100
Enable api compatibility test to also run on macos.
gunan Dec 18, 2017
d1801ac
Merge pull request #15454 from AndreiCostinescu/patch-3
yifeif Dec 18, 2017
1e99a68
Merge pull request #15386 from betterenvi/fix-gan-api-example
yifeif Dec 18, 2017
818a621
Merge pull request #15394 from qmick/patch-2
yifeif Dec 18, 2017
33b859a
Merge pull request #15459 from gunan/api
martinwicke Dec 18, 2017
300eab0
Merge pull request #15439 from snnn/m5
yifeif Dec 18, 2017
d71ab31
Merge pull request #15444 from snnn/fix1
yifeif Dec 18, 2017
9164903
merge
Dec 18, 2017
f3cc6cf
disable failing test
Dec 18, 2017
6d7a1bf
disable failing test, fix formatting
Dec 18, 2017
2411a72
Updating MKL to the latest release.
bhavani-subramanian Dec 19, 2017
bd13f3e
Merge pull request #15344 from javierluraschi/lite-readme-toco-link
yifeif Dec 19, 2017
44ce7f7
Add no_pip tags to stats_dataset_ops_test
yifeif Dec 19, 2017
fd9264c
`variable_scope` use `auxiliary_name_scope` to control whether to cre…
facaiy Dec 19, 2017
03dd228
Disable failing testcases on windows. (#15462)
gunan Dec 19, 2017
49b3b5e
Include _solib_local for MKL-DNN libs (#14709)
yingted Dec 19, 2017
1cc87d5
Merge pull request #15460 from ekelsen/branch_179464468
yifeif Dec 19, 2017
3ccdeda
Add decode_compressed support
yongtang Dec 5, 2017
b2904b0
Add DecodeCompressed ops to parsing_ops.cc
yongtang Dec 5, 2017
4fd57b8
Add test case for decode_compressed support
yongtang Dec 5, 2017
f1b8804
Add addtional test case to showcase combination of decode_compressed …
yongtang Dec 5, 2017
d5e285f
Add NONE ("") compression type support for decode_compressed
yongtang Dec 5, 2017
e95bfc5
Add test case for NONE ("") compression type with decode_compressed
yongtang Dec 5, 2017
6cfc88d
Buildifier tensorflow/core/kernels/BUILD
yongtang Dec 5, 2017
4f136c9
Update docstring.
yongtang Dec 18, 2017
8aa1b47
Fix python 3 error `a bytes-like object is required, not 'str'`
yongtang Dec 18, 2017
1434634
Update api_defs
yongtang Dec 19, 2017
24a52b6
TST: add test case
facaiy Dec 19, 2017
f08bce9
BUG: fix name scope collision in layer
facaiy Dec 19, 2017
5c4ecab
CLN: clean code
facaiy Dec 19, 2017
cef22b5
fix typos
ManHyuk Dec 19, 2017
5c0f165
fix issue building memory_stats with opencl
qmick Dec 19, 2017
8290e18
Fix a compile error in file_block_cache_test.cc
snnn Dec 19, 2017
a59863a
docstring
martinwicke Dec 19, 2017
96af950
Use Eigen version of the `scalar_fmod_op` for `fmod` ops (#15456)
yongtang Dec 19, 2017
ef8c5cf
Merge pull request #15476 from ManHyuk/fix_typo
yifeif Dec 19, 2017
196a9d3
Reverting changes in 69c324591ba4dfeafb403ee59de56ffe063c1e94 to fix …
sleighsoft Dec 19, 2017
5dc684f
Rerun tensorflow/core/api_def/update_api_def.sh
yongtang Dec 19, 2017
535d057
Merge commit for internal changes
Dec 19, 2017
dde5ca9
Merge pull request #15491 from ekelsen/branch_179578952
yifeif Dec 19, 2017
55cdadb
Merge pull request #15461 from Intel-tensorflow/mkl-0.11-update
yifeif Dec 19, 2017
6e96959
Merge pull request #15283 from nluehr/cudnn51-fix
yifeif Dec 19, 2017
74a5ff8
Merge pull request #14945 from blodan/master
yifeif Dec 19, 2017
6a40833
Merge pull request #15362 from akimach/patch-1
yifeif Dec 19, 2017
f962301
Merge pull request #15301 from rajendraarora16/patch-3
yifeif Dec 19, 2017
2bca9cd
Merge pull request #15422 from mikeknapp/patch-1
yifeif Dec 19, 2017
2728667
Merge pull request #15282 from nluehr/memory-stats-test-fix
yifeif Dec 19, 2017
25940a3
Merge pull request #15453 from AndreiCostinescu/patch-1
yifeif Dec 19, 2017
aafc38b
Merge pull request #14344 from cancan101/update_get_global_step
yifeif Dec 19, 2017
477d02a
Merge pull request #14330 from yongtang/14313-decode_libsvm
yifeif Dec 19, 2017
1bd5553
Merge pull request #15482 from snnn/fix6
yifeif Dec 19, 2017
f0e6c56
Merge pull request #15471 from facaiy/BUG/name_scope_collision_in_layer
yifeif Dec 19, 2017
7341e75
Merge pull request #15132 from yongtang/14887-decode_compressed
martinwicke Dec 20, 2017
1f43f12
Merge pull request #15479 from qmick/15477_fix_memory_stats_with_opencl
yifeif Dec 20, 2017
c619cff
Merge pull request #15474 from sleighsoft/TrainingHelperFix
yifeif Dec 20, 2017
28f24c1
Merge pull request #14370 from yongtang/11082017-conv1d-doc
martinwicke Dec 20, 2017
40e1c59
Merge pull request #13503 from dantkz/unstack_int64
yifeif Dec 20, 2017
18cf4ba
[XLA/tfcompile] Add Env::CreateTempFilename and use it in SaveGraph
rongjiecomputer Dec 19, 2017
b9fb50e
Address comments
rongjiecomputer Dec 20, 2017
5ab931b
Change name to CreateNewFileWithUniqueName
rongjiecomputer Dec 20, 2017
2c5876b
Revert clang-format
rongjiecomputer Dec 20, 2017
e7579f2
Change name to CreateUniqueFileName
rongjiecomputer Dec 20, 2017
1ad6f1b
Merge pull request #14638 from dmaclach/gitignore
yifeif Dec 20, 2017
fdae9d1
Merge pull request #14893 from powderluv/fix_xla_osx
yifeif Dec 20, 2017
f61c145
move to contrib
tjingrant Dec 20, 2017
ed2ec2d
add more comments
tjingrant Dec 20, 2017
53249a0
explain what SRU stands for
tjingrant Dec 20, 2017
c24f6dd
Update layers_test.py
shreyneil Dec 20, 2017
8207d25
fix typos
snnn Dec 20, 2017
4332891
[XLA] Use os.path for path manipulation
rongjiecomputer Dec 20, 2017
90a607a
[XLA] Add 'const' to custom comparator
rongjiecomputer Dec 20, 2017
be02bd1
[XLA] Explicitly include <numeric>
rongjiecomputer Dec 20, 2017
670a05e
[XLA] Use tensorflow::port::Aligned{Malloc,Free}
rongjiecomputer Dec 20, 2017
11278ba
[XLA] Fix std::array initialization
rongjiecomputer Dec 20, 2017
44432be
bugfix: long is 32 bits on Windows
snnn Dec 20, 2017
bc5ed11
[XLA] Hide GCC 7.1.1 workaround from MSVC
rongjiecomputer Dec 20, 2017
e83eeb5
Use simplified version of TF_ASSIGN_OR_RETURN for MSVC
rongjiecomputer Dec 20, 2017
dbf53d8
[XLA] Define LANG_CXX11 for >= VS 2015
rongjiecomputer Dec 20, 2017
4804eb3
fix typos
ManHyuk Dec 20, 2017
bac23c7
Remove LANG_CXX11 from get_win_copts
rongjiecomputer Dec 20, 2017
3424494
Clear the decref cache after calls to py_func to ensure no leaks
alextp Dec 20, 2017
710c8c9
Fix Golang readme installation instruction formatting
Dec 20, 2017
466a80f
Merge commit for internal changes
Dec 20, 2017
366ed06
Merge pull request #15523 from tensorflow/alextp-patch-1
yifeif Dec 20, 2017
f7616ae
C++ gradient for Select (#14862)
facaiy Dec 20, 2017
6120581
Merge pull request #15524 from uber/fix-golang-readme-formatting
yifeif Dec 20, 2017
bd8cc11
Merge pull request #15519 from ManHyuk/fix_typo
yifeif Dec 20, 2017
97bba41
Merge pull request #15504 from snnn/f901
yifeif Dec 20, 2017
b666042
[XLA] Fix XLA/tfcompile compile error on OSX
powderluv Dec 9, 2017
95e0469
Update debugger.md
kagemusha Dec 20, 2017
a1d2ac3
Merge pull request #15478 from rongjiecomputer/env
yifeif Dec 20, 2017
f48a045
Merge pull request #15229 from powderluv/fix_osx_tfcompile
yifeif Dec 20, 2017
6a415b0
Add no_pip tag to scan_dataset_op_test
yifeif Dec 20, 2017
9eacbf6
Merge pull request #15507 from rongjiecomputer/path
yifeif Dec 20, 2017
1902f04
Merge branch 'master' into master
yifeif Dec 20, 2017
67d9dca
Merge pull request #14673 from GreggHelt2/master
yifeif Dec 20, 2017
e82ec99
Merge pull request #14332 from pvaneck/jcenter-android
yifeif Dec 20, 2017
e944741
Merge pull request #15260 from ManHyuk/typo
yifeif Dec 20, 2017
642c925
Merge pull request #14527 from wodesuck/patch-1
yifeif Dec 20, 2017
78459af
Merge pull request #14478 from qmick/slice_op_remove_redundant
yifeif Dec 20, 2017
0f60bca
Merge pull request #15516 from rongjiecomputer/cxx
yifeif Dec 20, 2017
6fc96f0
Merge pull request #14377 from footh/layer-base-indent-fix
yifeif Dec 20, 2017
ae0a39a
Merge pull request #14403 from ZhengshengWei/r-2017-11-09
yifeif Dec 20, 2017
66b2f57
Merge pull request #14409 from tano297/jetson_platform
yifeif Dec 20, 2017
8fcbc68
Merge pull request #15512 from rongjiecomputer/hide_gcc_workaround
yifeif Dec 20, 2017
bde5837
Merge pull request #15511 from rongjiecomputer/array
yifeif Dec 21, 2017
9bddce9
Merge pull request #15509 from rongjiecomputer/numeric
yifeif Dec 21, 2017
0197101
Merge pull request #15508 from rongjiecomputer/const
yifeif Dec 21, 2017
21f38aa
[XLA] Define ssize_t for Windows
rongjiecomputer Dec 21, 2017
a4d4611
Merge pull request #14259 from vnavkal/svd
yifeif Dec 21, 2017
1ac7075
Remove old macro implementation
rongjiecomputer Dec 21, 2017
ab4abcd
Update comment
rongjiecomputer Dec 21, 2017
c50e95d
Merge pull request #15528 from kagemusha/patch-1
caisq Dec 21, 2017
9cba55c
Change sys.maxint to -1 in hook.py
yifeif Dec 21, 2017
fa17217
update sys.maxint to 1 in grpc_wrapper.py
yifeif Dec 21, 2017
542dce3
Use a trailing period to define floating point constant.
LucasSloan Dec 21, 2017
def58fb
Merge pull request #15536 from LucasSloan/fix-get-started-docs
yifeif Dec 21, 2017
c3eed95
Remove quantized_matmul_op_for_hexagon_test in Windows build scripts
snnn Dec 21, 2017
9920fd6
Merge pull request #15526 from ekelsen/branch_179628764
yifeif Dec 21, 2017
9a41aee
Updates imageops Docs
Dec 21, 2017
6475258
TST: unknown shape case
facaiy Dec 21, 2017
08a5db9
BUG: fix for unknown shape
facaiy Dec 21, 2017
b02e948
fix random_distributions_test
snnn Dec 21, 2017
e17bf69
golang: added Session.ListDevices method (#14385)
anight Dec 21, 2017
00aad3b
Merge with master.
theflofly Dec 21, 2017
5109eb5
Merge pull request #15539 from snnn/fix1
yifeif Dec 21, 2017
60e2fe3
Merge pull request #15515 from rongjiecomputer/macro
yifeif Dec 21, 2017
75a81f4
Merge pull request #15510 from rongjiecomputer/aligned_mem
yifeif Dec 21, 2017
a1e466d
Merge pull request #15436 from tjingrant/sru
yifeif Dec 21, 2017
17bdfb5
Merge pull request #15506 from snnn/fix10
yifeif Dec 21, 2017
149cad2
Merge commit for internal changes
yifeif Dec 21, 2017
11e3f35
Merge pull request #15550 from snnn/random_distributions_test
yifeif Dec 21, 2017
091f4fb
Merge pull request #15513 from rongjiecomputer/ssize
yifeif Dec 21, 2017
20db5c8
Merge pull request #15544 from facaiy/BUG/unknown_shape_for_sparse_mu…
yifeif Dec 21, 2017
b257445
Merge pull request #14422 from theflofly/add-gradient-prod-op
yifeif Dec 21, 2017
2242b67
fix typo
ManHyuk Dec 21, 2017
4e745b1
fix typo
ManHyuk Dec 21, 2017
62a9da3
Merge pull request #15562 from ManHyuk/fix_typo
yifeif Dec 21, 2017
060ac5c
Merge pull request #15545 from wagonhelm/updatedocs
yifeif Dec 21, 2017
09376cd
Merge pull request #15560 from yifeif/branch_179822007
yifeif Dec 21, 2017
5476407
Merge pull request #14785 from shreyneil/master
yifeif Dec 21, 2017
4549ec4
Adding MKL-DNN support for LRN op
claynerobison Dec 22, 2017
ab26d54
merge
Dec 22, 2017
1781c9f
Remove duplicate definition of compression_type
yifeif Dec 22, 2017
426164c
Merge pull request #15593 from ekelsen/branch_179953488
yifeif Dec 22, 2017
2df2d80
Update tf-learn reference to TF Estimators (#15596)
terrytangyuan Dec 23, 2017
9d9df8a
Merge pull request #15594 from Intel-tensorflow/pr-mkl-dnn-lrn
caisq Dec 24, 2017
6a86923
Merge commit for internal changes
caisq Dec 24, 2017
29086e5
Merge pull request #15615 from caisq/branch_180008567
caisq Dec 25, 2017
7da1738
Update: HTTP -> HTTPS
him2him2 Dec 25, 2017
c07d110
Reverting PR #14478, which breaks Inception v3 with MKL. (#15566)
claynerobison Dec 26, 2017
3c9b8b4
Merge commit for internal changes
Dec 26, 2017
cbeb616
Clean bazel `all_files` targets (#15368)
Androbin Dec 26, 2017
1ec58ed
add c++ gradient for op: Pow (#15245)
facaiy Dec 26, 2017
dd742c7
Merge pull request #15632 from drpngx/branch_180053468
caisq Dec 26, 2017
50aa27e
Merge pull request #15629 from him2him2/fix/HTTPS
caisq Dec 26, 2017
cd7fa34
Add support for CUBLAS_TENSOR_OP_MATH in fp16 GEMM (#13451)
nluehr Dec 26, 2017
d1ffdd4
Remove unneeded branch check (#13495)
yaroslavvb Dec 26, 2017
e085d10
Fixing MKL-DNN convolution filter propagation to backprop (#15227)
claynerobison Dec 26, 2017
6cfbd44
Update C API test data comparison for s390x (#15390)
shahidhs-ibm Dec 26, 2017
06b7ca1
Adding missing reorders in ReLU and AddN (#15406)
nhasabni Dec 26, 2017
620cf32
Add shape function SingleImageRandomDotStereograms (#15431)
yongtang Dec 26, 2017
ad0c2ea
call within the loop (#15446)
errord Dec 26, 2017
5683941
Add ws2_32.lib to gcs_dns_cache_test's linkopts (#15496)
snnn Dec 26, 2017
6362984
[XLA] Remove unused dlfcn.h and implement sincos[f] for MSVC (#15514)
rongjiecomputer Dec 26, 2017
4da4737
added note about weights gradient in compute_weighted_loss (#15533)
ngc92 Dec 26, 2017
5188b88
Bug fix for example_parsing_ops_test (#15548)
snnn Dec 26, 2017
a13f83e
Fix adding elements to collections.deque. (#15574)
yanboliang Dec 26, 2017
80232b2
[Bazel/MSVC] Fix build error since aae439c (#15579)
rongjiecomputer Dec 26, 2017
162104d
fix InvalidArgumentError when using cv2 with tf.py_func() (#14286)
EncodeTS Dec 26, 2017
eedc7d9
Merge pull request #14999 from n3011/patch-1
caisq Dec 26, 2017
82b8d48
Remove unwrapping for function argument inspection in Estimator. (#15…
drpngx Dec 27, 2017
b67873f
consistency in function name.. (#14260)
yupbank Dec 27, 2017
631dd5a
Add templated functions for `safe_strto[f|d|i32|i64|u32|u64]` (#14789)
yongtang Dec 27, 2017
bc05146
Add int64 support for BroadcastArgs and BroadcastGradientArgs (#15255)
yongtang Dec 27, 2017
a4abcff
Make the client_test able to be disabled using a manifest file (#15277)
DavidNorman Dec 27, 2017
3cde05e
Fix discrepancy between doc and impl for `tf.reverse` (#13672)
yongtang Dec 27, 2017
be4b688
New metric: Cohen's kappa (#15443)
facaiy Dec 27, 2017
2c82b8b
mpi_collectives: Refactor to fix build issues (#15534)
jthestness Dec 27, 2017
022d0f1
Recover the lost output channel number of atrous convolution for NC …
facaiy Dec 27, 2017
877876e
session_clusterspec_prop_test and common_runtime_direct_session_with_…
Dec 27, 2017
39f8090
Clarify batch_norm documentation to highlight that the dimensions for…
Dec 27, 2017
4db2463
Clarify batch_norm documentation to highlight that the dimensions for…
ZahlGraf Dec 27, 2017
d1c9a64
Merge branch 'clarify_batch_norm_documentation' of https://github.com…
ZahlGraf Dec 30, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
6 changes: 5 additions & 1 deletion tensorflow/contrib/layers/python/layers/layers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -479,7 +479,11 @@ def batch_norm(inputs,

Sergey Ioffe, Christian Szegedy

Can be used as a normalizer function for conv2d and fully_connected.
Can be used as a normalizer function for conv2d and fully_connected. The
dimensions on which the normalization is calculated depend on the tensor-shape.
In case of a 2D tensor the normalization is calculated over the first dimension
(batch) and in case of a 4D tensor the normalization is calculated over the
first three dimensions (batch and feature-map).

Note: when training, the moving_mean and moving_variance need to be updated.
By default the update ops are placed in `tf.GraphKeys.UPDATE_OPS`, so they
Expand Down