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.

Sign up for GitHub

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

Jump to bottom

Add uint32 and uint64 kernel support for Invert #15154

Merged
merged 4 commits into from
Feb 8, 2018
Merged

Add uint32 and uint64 kernel support for Invert #15154

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
6317915
Add uint32 and uint64 kernel support for `Invert`
yongtang Dec 5, 2017
b3ef18e
Add test cases for uint32 and uint64 support with `Invert`
yongtang Dec 5, 2017
2519b4e
Add missing uint32 and uint64 in GPU for invert
yongtang Dec 6, 2017
3a19416
Add DEFINE_UNARY8 for invert
yongtang Dec 6, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
2 changes: 1 addition & 1 deletion tensorflow/core/kernels/cwise_op_gpu_invert.cu.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@ limitations under the License.

namespace tensorflow {
namespace functor {
DEFINE_UNARY6(invert, int8, int16, int32, int64, uint8, uint16);
DEFINE_UNARY8(invert, int8, int16, int32, int64, uint8, uint16, uint32, uint64);
} // namespace functor
} // namespace tensorflow

Expand Down
10 changes: 5 additions & 5 deletions tensorflow/core/kernels/cwise_op_invert.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,17 +16,17 @@ limitations under the License.
#include "tensorflow/core/kernels/cwise_ops_common.h"

namespace tensorflow {
REGISTER6(UnaryOp, CPU, "Invert", functor::invert, int8, int16, int32, int64,
uint8, uint16);
REGISTER8(UnaryOp, CPU, "Invert", functor::invert, int8, int16, int32, int64,
uint8, uint16, uint32, uint64);

#ifdef TENSORFLOW_USE_SYCL
REGISTER6(UnaryOp, SYCL, "Invert", functor::invert, int8, int16, int32, int64,
uint8, uint16);
uint8, uint16, uint32, uint64);
#endif // TENSORFLOW_USE_SYCL

#if GOOGLE_CUDA
REGISTER6(UnaryOp, GPU, "Invert", functor::invert, int8, int16, int32, int64,
uint8, uint16);
REGISTER8(UnaryOp, GPU, "Invert", functor::invert, int8, int16, int32, int64,
uint8, uint16, uint32, uint64);
#endif // GOOGLE_CUDA

} // namespace tensorflow
3 changes: 3 additions & 0 deletions tensorflow/core/kernels/cwise_ops_gpu_common.cu.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -136,6 +136,9 @@ struct ApproximateEqual<GPUDevice, T> {
#define DEFINE_UNARY7(F, T0, T1, T2, T3, T4, T5, T6) \
DEFINE_UNARY2(F, T0, T1); \
DEFINE_UNARY5(F, T2, T3, T4, T5, T6)
#define DEFINE_UNARY8(F, T0, T1, T2, T3, T4, T5, T6, T7) \
DEFINE_UNARY4(F, T0, T1, T2, T3); \
DEFINE_UNARY4(F, T4, T5, T6, T7)

// Macros to explicitly instantiate kernels on GPU for multiple types
// (T0, T1, etc.) for BinaryFunctor.
Expand Down
2 changes: 1 addition & 1 deletion tensorflow/python/ops/bitwise_ops_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,7 +71,7 @@ def count_bits(x):

def testInvertOp(self):
dtype_list = [dtypes.int8, dtypes.int16, dtypes.int32, dtypes.int64,
dtypes.uint8, dtypes.uint16]
dtypes.uint8, dtypes.uint16, dtypes.uint32, dtypes.uint64]
inputs = [0, 5, 3, 14]
with self.test_session(use_gpu=True) as sess:
for dtype in dtype_list:
Expand Down