| Commit message (Collapse) | Author | Age |
|
|
|
|
|
| |
xla.estimator_model_fn
PiperOrigin-RevId: 214860981
|
|
|
|
| |
PiperOrigin-RevId: 214541221
|
|
|
|
| |
PiperOrigin-RevId: 214096165
|
|
|
|
| |
PiperOrigin-RevId: 214094266
|
|
|
|
|
|
| |
estimator API in order to compile entire model with XLA.
PiperOrigin-RevId: 214078470
|
|
|
|
|
|
|
|
| |
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: 213944355
|
|
|
|
| |
PiperOrigin-RevId: 213574904
|
|
|
|
| |
PiperOrigin-RevId: 210424333
|
|
|
|
|
|
| |
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
|
|
|
|
|
|
| |
This is in preparation for removing the _USE_C_API toggle altogether.
PiperOrigin-RevId: 196920481
|
|
|
|
|
|
|
| |
All integer tensors are now considered constant with respect to all `xs`.
This fixes a bug in gradients through tf.while_loop.
PiperOrigin-RevId: 194438529
|
|
|
|
| |
PiperOrigin-RevId: 190878279
|
|
|
|
| |
PiperOrigin-RevId: 184202470
|
|
|
|
| |
PiperOrigin-RevId: 172407754
|
|
|
|
| |
PiperOrigin-RevId: 165457067
|
|
|
|
| |
Change: 150460215
|
|
|
|
| |
Change: 149973410
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
blocks.
Gradients get their own separate scope based on the scope of the forward op.
Provide proper XlaScope for Defuns as well (each Defun gets its own scope; their gradients get their own scope).
Also move jit scope gradient unit tests out of core gradients to contrib.compiler.
This is just the python side that sets the attribute; the C++ changes will come
in a separate CL.
Change: 147216860
|
|
|
|
|
|
|
| |
experimental_jit_scope that can determine when an op should be compiled.
Add some unit tests of jit scope that check that compilation can be controlled using this new mechanism.
Change: 146034274
|
|
|
|
|
|
|
| |
XLA is a compiler-based linear algebra execution engine that targets CPUs, GPUs and custom accelerators.
XLA is still experimental; we are releasing it early to get the community involved.
Change: 143990941
|
|
|
|
| |
Change: 142318245
|
|
Change: 140860022
|