| f669885e1c135b1e5ad7b2e936083860a84b0aea |
|
10-Feb-2018 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Add python/util/is_in_graph_mode.py
PiperOrigin-RevId: 185260675
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| ec9592211fcd0940336c1f31dcac405709b6a247 |
|
27-Jan-2018 |
Martin Wicke <wicke@google.com> |
Adds a deprecated_alias utility function with which to deprecate unmodified aliases.
PiperOrigin-RevId: 183495796
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| b45c09ce715fb1d4d0a0f7e09586bd532031049e |
|
26-Jan-2018 |
Alexandre Passos <apassos@google.com> |
Improvements to eager linear regression benchmark:
1. Using _shape_tuple
2. Bypassing * over math_ops.mul etc
3. Flatmaps in the tape code
4. Cache for ones similar to for zeros
5. Fast path for _SubGrad
6. Fast global_step += 1 for resource variables
7. Bypassing deprecated args decorator in eager mode
PiperOrigin-RevId: 183446593
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 4982ef0fa4c2a2a2e9a438d42872425ec4ef5a0e |
|
13-Sep-2017 |
Martin Wicke <wicke@google.com> |
Add the ability to warn only once if deprecated functionality is used, and make that the default.
PiperOrigin-RevId: 168545655
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 325fc2330f4e527b394456ce1d821f2c73b1e7a2 |
|
23-May-2017 |
Martin Wicke <wicke@google.com> |
Make a silencing context manager for deprecations and use it.
PiperOrigin-RevId: 156870702
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| ee112cff56081fb9d0b74c987a8935acc360b05c |
|
11-May-2017 |
Benoit Steiner <bsteiner@google.com> |
Merge changes from github.
PiperOrigin-RevId: 155709893
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 8e5041918f2e709ded94e63fb1779d6bb363becb |
|
21-Apr-2017 |
Charles Nicholson <nicholsonc@google.com> |
Introduce TFDecorator, a base class for Python TensorFlow decorators. Provides basic introspection and "unwrap" services, allowing tooling code to fully 'understand' the wrapped object.
Change: 153854044
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 617e217cfcc2835752e839d3b6c1d28686b3a467 |
|
18-Apr-2017 |
Martin Wicke <wicke@google.com> |
Allow date=None for deprecation warnings, which will render as "in a future version".
Change: 153470458
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| d1ba01f81d8fa1d0171ba9ce871599063d5c7eb9 |
|
02-Feb-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Merge changes from github.
Change: 146316196
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 5462618b916cad9aec46a9d1fd04ef353e9696b9 |
|
15-Dec-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Mutes warnings for some deprecated arguments with default values. Fixes a bug.
Users were getting log warnings even when using the API correctly,
because of argument passthrough within tf.learn. This change mutes
warnings for calls to fit, estimate, and predict if the deprecated
arguments receive default values.
Also fixes a bug in the deprecation tool where "==" was used instead of "is" to compare objects, which was at times yeilding a non-boolean.
Change: 142162193
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| db7305f448a8ca3e7a7b53a03121c5197b52e658 |
|
12-Dec-2016 |
Andrew Selle <aselle@google.com> |
Speedup deprecation printing 1250x
Avoid inspect.stack() which is super slow. Instead, use inspect.currentframe()
when CPython is available. This is what logging uses (albeit accessing
with a protected sys method)
Change: 141779089
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 83cef3f1f4ce4177c5bcf38b6a24e920dc12fba7 |
|
01-Dec-2016 |
Andrew Selle <aselle@google.com> |
Rename more argument names to be consistent with NumPy.
Create temporary dummy arguments with old names at the end in preparation for
deprecation.
Change: 140675930
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 4bc1d1cece4df20fdd56303276c637df97dcbeb7 |
|
28-Nov-2016 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Modifies deprecated_args to not warn if value == ok_value.
tf.learn currently has a problem wherein the user is warned even if
they are using the preferred API. This is because of arguments being
passed through API layers. This change expands the deprecation API
by allowing the deprecator to specify "ok_values" for which warnings
will be suppressed. Specifying ok_values for certain variables, such
as ("x", None) should reduce log warning spam in tf.learn.
Change: 140379168
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 12a53b859b4648dbebfc823b41cc3d61ba49cfda |
|
04-Nov-2016 |
Illia Polosukhin <ipolosukhin@google.com> |
Updated deprecation module to print location of call to deprecated function/arguments to make it easy to find. Updated deprecated_args to print which args exist if wrong argument was passed.
Change: 138201843
/external/tensorflow/tensorflow/python/util/deprecation.py
|
| 6bd7f4d7d7866db1d1406266f25d656bc5f54364 |
|
01-Nov-2016 |
Illia Polosukhin <ipolosukhin@google.com> |
Move deprecation module into core framework from contrib/framework
Change: 137850119
/external/tensorflow/tensorflow/python/util/deprecation.py
|