| Commit message (Collapse) | Author | Age |
|
|
|
| |
PiperOrigin-RevId: 216443201
|
|
|
|
| |
PiperOrigin-RevId: 215239710
|
|
|
|
| |
PiperOrigin-RevId: 215025019
|
|
|
|
| |
PiperOrigin-RevId: 215010842
|
|
|
|
|
|
|
|
| |
This change reduce the size of //tensorflow/tools/pip_package:simple_console_windows's zip file from 1000027677 bytes to 47690474 bytes for a CPU build. For GPU build, it will avoid going over 4GB when multiple CUDA compatibility are specified.
To fix #22390
PiperOrigin-RevId: 214764423
|
|\ |
|
| |
| |
| |
| | |
PiperOrigin-RevId: 214091820
|
| |
| |
| |
| |
| |
| |
| |
| | |
self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about:
* the fact that the session may be reused.
* the session is not closed even when doing a "with self.test_session()" statement.
PiperOrigin-RevId: 212725342
|
| |
| |
| |
| |
| |
| | |
it more accurate.
PiperOrigin-RevId: 212555968
|
| |
| |
| |
| |
| |
| | |
The fix is straightforward enough, although the triggering circumstances are still a bit mysterious. The unit test did fail with ubsan prior to this CL, so I'm going to leave it at that for now.
PiperOrigin-RevId: 212465732
|
| |
| |
| |
| |
| |
| |
| |
| | |
self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about:
* the fact that the session may be reused.
* the session is not closed even when doing a "with self.test_session()" statement.
PiperOrigin-RevId: 212348850
|
| |
| |
| |
| | |
PiperOrigin-RevId: 210407945
|
| |
| |
| |
| |
| |
| | |
Apparently it's not inherited from its py_binary dep for pip tests
PiperOrigin-RevId: 209848894
|
| |
| |
| |
| |
| |
| | |
self.test_session() has been deprecated in cl/208545396 as its behavior confuses readers of the test. Moving to self.session() instead.
PiperOrigin-RevId: 209696110
|
| |
| |
| |
| |
| |
| | |
Doesn't require an input_filename flag, uses the default example input instead.
PiperOrigin-RevId: 209518060
|
| |
| |
| |
| | |
PiperOrigin-RevId: 209498358
|
| |
| |
| |
| |
| |
| | |
I saw a MacOS flake after cl/208297005. Something platform-dependent seems to be going on.
PiperOrigin-RevId: 209220306
|
| |
| |
| |
| | |
PiperOrigin-RevId: 208297005
|
| |
| |
| |
| | |
PiperOrigin-RevId: 208126204
|
| |
| |
| |
| |
| |
| | |
It doesn't need to be fed (OneShot), and returning it as a tuple causes some issues for evaluation infrastructure.
PiperOrigin-RevId: 207820650
|
| |
| |
| |
| |
| |
| | |
Works with the one-shot head (no model state in the tf.Example proto).
PiperOrigin-RevId: 206988925
|
| |
| |
| |
| |
| |
| | |
Apparently lists and tuples behave differently in some numpy versions.
PiperOrigin-RevId: 206616759
|
| |
| |
| |
| | |
PiperOrigin-RevId: 203806903
|
| |
| |
| |
| |
| |
| | |
Adds a unit test for OneShotHead, fiddles with the train_op to deal with a dtype error the new unit tests uncovered.
PiperOrigin-RevId: 203179477
|
| |
| |
| |
| | |
PiperOrigin-RevId: 202115471
|
| |
| |
| |
| |
| |
| | |
Relnotes: hooks will now see deterministically the value of the global
step before updating instead of the value after updating.
PiperOrigin-RevId: 202000826
|
| |
| |
| |
| |
| |
| | |
evaluation op.
PiperOrigin-RevId: 200747192
|
| |
| |
| |
| | |
PiperOrigin-RevId: 197097430
|
| |
| |
| |
| |
| |
| | |
Replaces a "loss decreased" check with basic shape checking (it should have been seeded already, so there's likely some race condition which I should track down...).
PiperOrigin-RevId: 196001526
|
| |
| |
| |
| |
| |
| | |
opt mode to avoid flaky timeouts
PiperOrigin-RevId: 195993828
|
| |
| |
| |
| |
| |
| |
| |
| | |
Adds a very simple LSTM encoder/decoder option as an example.
ARModel's new constructor argument is a bit awkward, since Estimator's new graphs mean we need a Model factory rather than a Model (or to un-build the model?). It's still a much more pleasant way to write autoregressive models than fiddling with ARModel directly, since ARModel handles collecting all the features (and the prediction loop, etc.). Happy to hear other ideas for an API.
PiperOrigin-RevId: 195436186
|
| |
| |
| |
| |
| |
| | |
tensorflow/contrib/timeseries/python/timeseries/state_space_models:structural_ensemble_test
PiperOrigin-RevId: 195019968
|
| |
| |
| |
| | |
PiperOrigin-RevId: 194997009
|
| |
| |
| |
| | |
PiperOrigin-RevId: 194031845
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
They get flattened with the endogenous features as input to the model. Unlike
endogenous features, they're specified for the whole window when making
predictions.
Adds an ARRegressor example which uses exogenous features.
PiperOrigin-RevId: 194006630
|
| |
| |
| |
| | |
PiperOrigin-RevId: 193246563
|
|/ |
|
|
|
|
| |
PiperOrigin-RevId: 192388250
|
|
|
|
|
|
| |
Disabling the value-based check for now. Hopefully the shapes are deterministic.
PiperOrigin-RevId: 192383553
|
|
|
|
|
|
| |
This reverts commit 4e108ef30d7cd7ae5e1c550ec5ae27e79b8c6e39.
PiperOrigin-RevId: 191391075
|
|
|
|
| |
PiperOrigin-RevId: 191373516
|
|
|
|
|
|
| |
Removes state where it wasn't used.
PiperOrigin-RevId: 191324834
|
|
|
|
| |
PiperOrigin-RevId: 190953197
|
|
|
|
| |
PiperOrigin-RevId: 190878279
|
|
|
|
| |
PiperOrigin-RevId: 190858242
|
|
|
|
| |
PiperOrigin-RevId: 190835392
|
|
|
|
|
|
|
| |
It now correctly broadcasts start state across whatever batch dimension it is
passed rather than sqishing it down to a batch dimension of 1.
PiperOrigin-RevId: 190688855
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This means the model starts from its default start state and is fed a series
(filtering) to warm up its state. This warmed up state can then be used to
make predictions.
Some shape fiddling with the receiver_fn to make feeding state optional, and a
new signature for cold-starting which uses the model's default start state.
Some other shape fiddling to make feeding strings to SavedModels work more
smoothly in the cold-start part of the LSTM example. I was squeezing out the
last dimension of "scalar" exogenous features, now I'm leaving them, which
matches the placeholder generation logic.
PiperOrigin-RevId: 189414869
|
|
|
|
| |
PiperOrigin-RevId: 189258641
|
|
|
|
| |
PiperOrigin-RevId: 189231636
|