| .. |
|
arithmetic_test.cc
|
[XLA] Implement stable argmin and argmax
|
2020-04-18 00:09:30 -07:00 |
|
arithmetic.cc
|
Avoid use of ValueOrDie without inspecting the error in compiler/xla/client
|
2020-08-13 13:34:31 -07:00 |
|
arithmetic.h
|
[XLA] Implement stable argmin and argmax
|
2020-04-18 00:09:30 -07:00 |
|
BUILD
|
Add target environment constraints to a subset of TensorFlow packages and targets.
|
2020-09-21 10:57:01 -07:00 |
|
comparators_test.cc
|
|
|
|
comparators.cc
|
Avoid use of ValueOrDie without inspecting the error in compiler/xla/client
|
2020-08-13 13:34:31 -07:00 |
|
comparators.h
|
Adding total-order comparison support in proto and HloInstruction.
|
2020-08-10 09:40:29 -07:00 |
|
constants_test.cc
|
|
|
|
constants.cc
|
Replace tensorflow::bfloat16 with Eigen::bfloat16
|
2020-07-01 11:13:16 +08:00 |
|
constants.h
|
|
|
|
conv_grad_size_util.cc
|
|
|
|
conv_grad_size_util.h
|
|
|
|
logdet_test.cc
|
[XLA:CLIENT] Make LogDet work for non positive-definite matrices by transforming the matrix with multiplication by its conjugate transpose.
|
2020-08-25 11:30:56 -07:00 |
|
logdet.cc
|
[XLA:CLIENT] Make LogDet work for non positive-definite matrices by transforming the matrix with multiplication by its conjugate transpose.
|
2020-08-25 11:30:56 -07:00 |
|
logdet.h
|
|
|
|
loops.cc
|
|
|
|
loops.h
|
|
|
|
lu_decomposition.cc
|
[XLA:TPU] Add a TPU implementation of LU decomposition using a CustomCall.
|
2020-09-15 04:30:19 -07:00 |
|
lu_decomposition.h
|
[XLA:TPU] Add a TPU implementation of LU decomposition using a CustomCall.
|
2020-09-15 04:30:19 -07:00 |
|
math_test.cc
|
Add complex data type support for tf.math.acos in XLA
|
2020-07-14 18:47:03 +00:00 |
|
math.cc
|
[XLA] Switch implementation of erf to use the same rational polynomial approximation as Eigen.
|
2020-10-15 11:29:25 -07:00 |
|
math.h
|
Add XLA registration for Polygamma, to be used for Digamma gradients.
|
2020-09-18 17:40:20 -07:00 |
|
matrix_test.cc
|
[XLA:CLIENT] Support all gradients of <= 2 operand einsums
|
2020-08-24 11:06:18 -07:00 |
|
matrix.cc
|
[XLA:CLIENT] The reduces in Einsum can occur at the inputs instead of the
|
2020-08-26 14:11:31 -07:00 |
|
matrix.h
|
[XLA:CLIENT] Support all gradients of <= 2 operand einsums
|
2020-08-24 11:06:18 -07:00 |
|
pooling_test.cc
|
|
|
|
pooling.cc
|
Merge pull request #41752 from tg-at-google:wsign-compare-semi-final-xla
|
2020-07-29 19:10:26 -07:00 |
|
pooling.h
|
|
|
|
prng.cc
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
prng.h
|
Internal change
|
2020-09-17 21:37:45 -07:00 |
|
qr_test.cc
|
Use TensorFloat-32 instead of TF32 in comments and symbols.
|
2020-09-08 17:38:28 -07:00 |
|
qr.cc
|
[XLA] Change the QR decomposition implementation to lower to a CustomCall operator that is later expanded by a new QrExpander pass. This allows backends to lower the Qr decomposition differently if they choose.
|
2020-09-21 17:35:34 -07:00 |
|
qr.h
|
|
|
|
quantize_test.cc
|
|
|
|
quantize.h
|
Move bfloat16 header to tensorflow/core/platform.
|
2020-08-14 13:44:57 -07:00 |
|
self_adjoint_eig_test.cc
|
|
|
|
self_adjoint_eig.cc
|
Avoid use of ValueOrDie without inspecting the error in compiler/xla/client
|
2020-08-13 13:34:31 -07:00 |
|
self_adjoint_eig.h
|
|
|
|
slicing_test.cc
|
Propagate test error to avoid crashes.
|
2020-09-15 14:23:50 -07:00 |
|
slicing.cc
|
xla directory resolutions
|
2020-07-26 22:14:33 +00:00 |
|
slicing.h
|
|
|
|
sorting_test.cc
|
Call TopKWithPartition when input is large.
|
2020-08-27 22:33:34 -07:00 |
|
sorting.cc
|
Increase threshold for K to triggered TopKWithPartition.
|
2020-09-09 20:19:13 -07:00 |
|
sorting.h
|
|
|
|
svd_test.cc
|
|
|
|
svd.cc
|
Avoid use of ValueOrDie without inspecting the error in compiler/xla/client
|
2020-08-13 13:34:31 -07:00 |
|
svd.h
|
|
|
|
testing.cc
|
|
|
|
testing.h
|
|
|
|
tridiagonal_test.cc
|
Unlock XLA non pivoting tridiagonal solver for all back-ends.
|
2020-03-23 17:23:42 -07:00 |
|
tridiagonal.cc
|
Unlock XLA non pivoting tridiagonal solver for all back-ends.
|
2020-03-23 17:23:42 -07:00 |
|
tridiagonal.h
|
|
|
