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.

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

Branch 141962539 #6296

Merged
merged 112 commits into from Dec 14, 2016
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
112 commits
Select commit Hold shift + click to select a range
a3f839e
Support TransformedDistribution entropy in the limited case when bije…
jvdillon Dec 9, 2016
71b28db
Update ops-related pbtxt files.
tensorflower-gardener Dec 9, 2016
66907af
Change references to TensorArray.pack, TensorArray.unpack to TensorAr…
tensorflower-gardener Dec 9, 2016
6d400d4
Remove some of the deprecations added to logging_ops.py in cl/1415159…
tensorflower-gardener Dec 9, 2016
c49e52c
Remove an unnecessary assert in metrics.streaming_sparse_precision_at…
tensorflower-gardener Dec 9, 2016
760803a
Prevent ability to take second derivs of ctc_loss_op until it is poss…
Dec 9, 2016
b48ec9a
Add missing test rule for draw_bounding_box_op
jart Dec 9, 2016
da9f662
Add missing test for adagrad_da
jart Dec 9, 2016
996d477
boolean_mask BUGFIX.
langmore Dec 9, 2016
d499f52
Update generated Python Op docs.
tensorflower-gardener Dec 9, 2016
1da7d8a
Fix documentation link in SavedModel.
Dec 10, 2016
4cffce0
Add missing test rule for registry_test
jart Dec 10, 2016
e390dcd
Automated rollback of change 140909396
vinuraja Dec 10, 2016
892dc07
Update Bazel version to 0.4.2
jart Dec 10, 2016
6da1cfa
Change references to TensorArray.pack, TensorArray.unpack to TensorAr…
tensorflower-gardener Dec 10, 2016
35a58c8
Change references to TensorArray.pack, TensorArray.unpack to TensorAr…
tensorflower-gardener Dec 10, 2016
92cc148
Name the nameless build rules
jart Dec 10, 2016
047072a
Update generated Python Op docs.
tensorflower-gardener Dec 10, 2016
f327050
Undo the changes in python for the PreventGradient addition - adding the
Dec 10, 2016
1fa6bc3
In sparse_to_dense_op.cc, include eigen before other libraries.
tensorflower-gardener Dec 10, 2016
7bd9a1c
Use redundant download URLs
jart Dec 10, 2016
d8f2a4b
Change references to TensorArray.pack, TensorArray.unpack to TensorAr…
tensorflower-gardener Dec 10, 2016
13b0c97
Update caller to move from tf.nn.rnn to (the identical) tf.contrib.rn…
Dec 10, 2016
6c8fbf6
tfdbg: fix a dtype-related bug in debug_data.is_inf_or_nan
caisq Dec 11, 2016
5d29db0
Dedup multiple control edges from one partition to a single destinati…
yuanbyu Dec 11, 2016
6214c59
Unconditionally invoke the function that removes external control edg…
yuanbyu Dec 11, 2016
29c17b6
Revert breaking changes from #5546
davidsoergel Dec 11, 2016
86afb78
Switch tf.Learn to CheckpointV2 and fix tests.
davidsoergel Dec 11, 2016
ff16c6e
Update generated Python Op docs.
tensorflower-gardener Dec 11, 2016
ed814ca
tfdbg: add debug op DebugNumericSummary
caisq Dec 12, 2016
627f64a
Update ops-related pbtxt files.
tensorflower-gardener Dec 12, 2016
d45ccfc
Remove leftover file.
gunan Dec 12, 2016
4ae96e5
changing zeros_initializer API to be like other callable initializers
itsmeolivia Dec 12, 2016
2879eae
Update generated Python Op docs.
tensorflower-gardener Dec 12, 2016
ed62a18
Remove hourglass import from python/debug tests
jart Dec 12, 2016
1943a9b
Merge TensorForest ops into one library/generated python file, gettin…
tensorflower-gardener Dec 12, 2016
684e62b
Disable antialiasing, implementation across browsers makes it impossi…
Dec 12, 2016
9af0e2b
Automated rollback of change 141521124
martinwicke Dec 12, 2016
d204d0f
Create tf.reverse() to match the behavior of tf.reverse_v2().
aselle Dec 12, 2016
db7305f
Speedup deprecation printing 1250x
aselle Dec 12, 2016
241be8f
Update ops-related pbtxt files.
tensorflower-gardener Dec 12, 2016
588fc71
Update generated Python Op docs.
tensorflower-gardener Dec 12, 2016
2e47277
Changes to zeros_initializer which seem to have been missed in the mi…
tensorflower-gardener Dec 12, 2016
573d8ca
Registers resource handles by default for most kernels.
tensorflower-gardener Dec 12, 2016
2eaaada
Hashing function input/output args should be a part of the function
tensorflower-gardener Dec 12, 2016
552a641
Remove deprecated weight and target args.
tensorflower-gardener Dec 12, 2016
6e479d3
Refactor unit tests to make them easier to follow.
tensorflower-gardener Dec 12, 2016
ad7d715
Android: minor update to README.md specifying Activity labels.
andrewharp Dec 12, 2016
db979cd
Update generated Python Op docs.
tensorflower-gardener Dec 12, 2016
dc60657
Squeezing labels before passing them into the top k eval metric for T…
tensorflower-gardener Dec 12, 2016
68433db
Add TrackingVisitableAllocator subclass of TrackingAllocator and Visi…
tensorflower-gardener Dec 12, 2016
c0047ab
Allow last dimension of labels for "at_k" metrics to be missing if nu…
tensorflower-gardener Dec 12, 2016
65a5a35
Fix typo in debugger howto.
dm-jrae Dec 12, 2016
5343d03
Fix typo in master_service.proto documentation
saeta Dec 12, 2016
2166d20
Before the embedding projector loads, check if WebGL is enabled (this…
Dec 12, 2016
e86fce7
Update generated Python Op docs.
tensorflower-gardener Dec 12, 2016
61eff53
Two groups of changes applied to tensorflow_models/tutorials:
nealwu Dec 12, 2016
5573537
Add string_to_index_table, which returns a lookup table that matches …
ysuematsu Dec 12, 2016
a46b6d2
Major intent of this CL is to rename split_v -> split in the python API.
tensorflower-gardener Dec 12, 2016
6955d56
Automatically open the bookmarks tray if there are bookmarks when swi…
Dec 12, 2016
e43b4e4
Update generated Python Op docs.
tensorflower-gardener Dec 12, 2016
e0587d1
Script to build an optimized version of the protobuf package in CI.
Dec 12, 2016
1294031
Fix t-sne bug (division by 0) when there are duplicate points.
Dec 12, 2016
80d283f
Add method to signal the hooks of a recovered session.
adria-p Dec 12, 2016
43b29aa
Release new version of TensorBoard to OSS.
Dec 13, 2016
839dd00
Changes to `Affine`: Make `shift` an optional argument. Make event_nd…
jvdillon Dec 13, 2016
62dae8b
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
db00a72
(Part 1 of 2.) Add arguments to TransformedDistribution which allow i…
jvdillon Dec 13, 2016
5d43d55
Avoid both conditional branches being computed in Dropout and BatchNo…
fchollet Dec 13, 2016
7902dc1
Remove links to TensorBoard demo
jart Dec 13, 2016
7f38c5f
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
3e524a3
Change implementation of concat grad to use new split. Results in up…
tensorflower-gardener Dec 13, 2016
f6585e1
Replace saved_model_half_plus_two_data genrule with SavedModel golden…
lilao Dec 13, 2016
b714503
Refactor StatSummarizer and make it print a summary by node type.
andrewharp Dec 13, 2016
f3a1f0a
Add deprecation annotations to tf.pack, tf.unpack and tf.concat.
tensorflower-gardener Dec 13, 2016
70bcbf3
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
b6ab735
Update caller to move from tf.nn.rnn to (the identical) tf.contrib.rn…
Dec 13, 2016
f8e7878
Update caller to move from tf.nn.rnn to (the identical) tf.contrib.rn…
Dec 13, 2016
77d641c
Automated rollback of change 141555939
ringw Dec 13, 2016
42376e9
Replace SessionBundle half_plus_two genrule with export golden files.
lilao Dec 13, 2016
a467759
Swap the order of inputs and labels of the sampled_softmax_loss.
Dec 13, 2016
c02b453
Added index_to_string_table which returns a lookup table that maps in…
ysuematsu Dec 13, 2016
04f073a
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
c0076de
Rename SyncReplicasOptimizerV2 to SyncReplicasOptimizer
tensorflower-gardener Dec 13, 2016
780bc47
Update caller to move from tf.nn.rnn to (the identical) tf.contrib.rn…
Dec 13, 2016
f4e947c
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
40b2061
Add contrib.opt to TF documentation.
tensorflower-gardener Dec 13, 2016
41a48ab
Add ConcatV2 test helper method
freeblee Dec 13, 2016
af318c9
Fix MonitoredSession markdown formatting.
Dec 13, 2016
6191fdc
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
492c397
Bump up size of bijector_test.
tensorflower-gardener Dec 13, 2016
a26aad7
Add numpy style .shape to Tensor class...
tensorflower-gardener Dec 13, 2016
0085d31
For extra safety, add `abs` to `log` calls in `OperatorPD` `det` func…
jvdillon Dec 13, 2016
6515cfd
Update generated Python Op docs.
tensorflower-gardener Dec 13, 2016
63fbdc8
Add unit tests for compute_weighted_loss.
tensorflower-gardener Dec 13, 2016
2b771b9
Fix use-after-free in the DirectSession timeout handling code.
mrry Dec 13, 2016
df5d72d
Update comment to match current kernel registration macro.
tensorflower-gardener Dec 13, 2016
1bd4c19
Remove stale files.
yifeif Dec 13, 2016
0b217fc
Decompose state into multiple tensors and return them in prediction d…
tensorflower-gardener Dec 13, 2016
f9e1a91
Add sparse_column_with_vocabulary_file, to specify a feature column t…
ysuematsu Dec 13, 2016
fbc8107
LinearOperatorComposition: Add a test for the name.
langmore Dec 13, 2016
9c390e1
tfdbg CLI: add bash-style output-to-file support
caisq Dec 13, 2016
caaf688
Optimize the case of a client communicating with an in-process master.
mrry Dec 13, 2016
f0a6d1e
Release notes update
aselle Dec 14, 2016
667a023
Find libdevice files in .runfiles when TEST_SRCDIR is not set.
Dec 14, 2016
8d05819
Fix slowdown in graph building introduced by repeated calls to `inspe…
fchollet Dec 14, 2016
2f0cd86
Internal only changes
gunan Dec 14, 2016
527556c
Merge BUILD file changes
martinwicke Dec 14, 2016
c236f27
Fix linter issues
martinwicke Dec 14, 2016
06ae1e2
Fix initializer dtype problem
martinwicke Dec 14, 2016
5a1498a
Fix initializer dtype problem
martinwicke Dec 14, 2016
bc04100
Add tensor_forest to cmake build
martinwicke Dec 14, 2016
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
47 changes: 44 additions & 3 deletions RELEASE.md
Expand Up @@ -3,9 +3,50 @@
## Breaking Changes to the API

* Division and modulus operators (/, //, %) now match Python (flooring)
semantics. tf.div is renamed to tf.division. New operators tf.truncatediv and
tf.truncatemod are available for achieving the previous C++ (truncation)
division/modulus semantics.
semantics. This applies to `tf.div` and `tf.mod` as well. To obtain forced
integer truncation based behaviors you can use `tf.truncatediv`
and `tf.truncatemod`.
* `tf.divide()` is now the recommended division function. `tf.div()` will
remain, but its semantics do not respond to Python 3 or `from future`
mechanisms.
* tf.reverse() now takes indices of axes to be reversed. E.g.
`tf.reverse(a, [True, False, True])` must now be written as
`tf.reverse(a, [0, 2])`. `tf.reverse_v2()` will remain until 1.0 final.
* `tf.mul`, `tf.sub` and `tf.neg` are deprecated in favor of `tf.multiply`,
`tf.subtract` and `tf.negative`.
* The following Python functions have had their arguments changed to use `axis`
when referring to specific dimensions. We have kept the old keyword arguments
for compatibility currently, but we will be removing them well before the
final 1.0.
* `tf.argmax`: `dimension` becomes `axis`
* `tf.argmin`: `dimension` becomes `axis`
* `tf.count_nonzero`: `reduction_indices` becomes `axis`
* `tf.expand_dims`: `dim` becomes `axis`
* `tf.reduce_all`: `reduction_indices` becomes `axis`
* `tf.reduce_any`: `reduction_indices` becomes `axis`
* `tf.reduce_join`: `reduction_indices` becomes `axis`
* `tf.reduce_logsumexp`: `reduction_indices` becomes `axis`
* `tf.reduce_max`: `reduction_indices` becomes `axis`
* `tf.reduce_mean`: `reduction_indices` becomes `axis`
* `tf.reduce_min`: `reduction_indices` becomes `axis`
* `tf.reduce_prod`: `reduction_indices` becomes `axis`
* `tf.reduce_sum`: `reduction_indices` becomes `axis`
* `tf.reverse_sequence`: `batch_dim` becomes `batch_axis`, `seq_dim` becomes `seq_axis`
* `tf.sparse_concat`: `concat_dim` becomes `axis`
* `tf.sparse_reduce_sum`: `reduction_axes` becomes `axis`
* `tf.sparse_reduce_sum_sparse`: `reduction_axes` becomes `axis`
* `tf.sparse_split`: `split_dim` becomes `axis`
* `tf.listdiff` has been renamed to `tf.setdiff1d` to match NumPy naming.
* `tf.inv` has been renamed to be `tf.reciprocal` (component-wise reciprocal)
to avoid confusion with `np.inv` which is matrix inversion
* tf.round now uses banker's rounding (round to even) semantics to match NumPy.
* `tf.split` now takes arguments in a reversed order and with different
keywords. In particular, we now match NumPy order as
`tf.split(value, num_or_size_splits, axis)`.
* `tf.sparse_split` now takes arguments in reversed order and with different
keywords. In particular we now match NumPy order as
`tf.sparse_split(sp_input, num_split, axis)`. NOTE: we have temporarily
made `tf.sparse_split` require keyword arguments.

# Release 0.12.0

Expand Down
2 changes: 1 addition & 1 deletion WORKSPACE
Expand Up @@ -20,7 +20,7 @@ tf_workspace()

# Specify the minimum required bazel version.
load("//tensorflow:tensorflow.bzl", "check_version")
check_version("0.3.2")
check_version("0.4.2")

new_http_archive(
name = "inception5h",
Expand Down
50 changes: 29 additions & 21 deletions tensorflow/cc/saved_model/BUILD
Expand Up @@ -41,27 +41,35 @@ cc_library(
],
)

# Reenable target once half_plus_two_data does not run TF during build.
# tf_cc_test(
# name = "loader_test",
# srcs = ["loader_test.cc"],
# data = [
# "//tensorflow/python/saved_model/example:saved_model_half_plus_two_data",
# ],
# linkstatic = 1,
# tags = ["manual"],
# deps = [
# ":constants",
# ":loader",
# ":signature_constants",
# ":tag_constants",
# "//tensorflow/core:lib",
# "//tensorflow/core:protos_all_cc",
# "//tensorflow/core:test",
# "//tensorflow/core:test_main",
# "//tensorflow/core:testlib",
# ],
# )
tf_cc_test(
name = "loader_test",
srcs = ["loader_test.cc"],
data = [
":saved_model_half_plus_two",
],
linkstatic = 1,
tags = ["manual"],
deps = [
":constants",
":loader",
":signature_constants",
":tag_constants",
"//tensorflow/core:lib",
"//tensorflow/core:protos_all_cc",
"//tensorflow/core:test",
"//tensorflow/core:test_main",
"//tensorflow/core:testlib",
],
)

# TODO(b/32673259): add a test to continuously validate these files.
filegroup(
name = "saved_model_half_plus_two",
srcs = glob([
"testdata/half_plus_two_pbtxt/**",
"testdata/half_plus_two/**",
]),
)

# -----------------------------------------------------------------------------
# Google-internal targets.
Expand Down
4 changes: 2 additions & 2 deletions tensorflow/cc/saved_model/loader_test.cc
Expand Up @@ -30,9 +30,9 @@ namespace tensorflow {
namespace {

constexpr char kTestDataPbTxt[] =
"python/saved_model/example/saved_model_half_plus_two_pbtxt/00000123";
"cc/saved_model/testdata/half_plus_two_pbtxt/00000123";
constexpr char kTestDataSharded[] =
"python/saved_model/example/saved_model_half_plus_two/00000123";
"cc/saved_model/testdata/half_plus_two/00000123";

class LoaderTest : public ::testing::Test {
protected:
Expand Down
@@ -0,0 +1 @@
asset-file-contents
Binary file not shown.
Binary file not shown.
Binary file not shown.
@@ -0,0 +1 @@
asset-file-contents