diff options
| author |  Peter Hawkins <phawkins@google.com> | 2017-10-10 08:36:23 -0700 |
|---|---|---|
| committer |  TensorFlower Gardener <gardener@tensorflow.org> | 2017-10-10 08:39:56 -0700 |
| commit | 3bafe0a86f67dd54197c6d60bdb5053f510de7d8 (patch) | |
| tree | b7e9b7fd922f1c5d96ebf04465cfb45154964b4c /tensorflow/go/tensor.go | |
| parent | 5a26d1ede506825455d1199267d88caeba7d206a (diff) | |
Add uint32 and uint64 types to TensorFlow.
This change merely creates the types, but does not register kernels that act on uint32/uint64 values. It also does not alter most op registration lists to include uint32/uint64 values. If desirable, that can be done in a subsequent change, although binary size will likely prove problematic if adding more kernels.
The intent of the change is so XLA-compiled code can make use uint32/uint64 types. Since XLA does not use traditional TensorFlow kernels, using uint32/uint64 operators from XLA will require only uint32/uint64 op registrations, but will require few new kernel registrations.
PiperOrigin-RevId: 171681867
Diffstat (limited to 'tensorflow/go/tensor.go')
| -rw-r--r-- | tensorflow/go/tensor.go | 4 |
1 files changed, 4 insertions, 0 deletions
diff --git a/tensorflow/go/tensor.go b/tensorflow/go/tensor.go index e8fa21a62b..36a74c0081 100644 --- a/tensorflow/go/tensor.go +++ b/tensorflow/go/tensor.go @@ -39,6 +39,7 @@ const ( Float DataType = C.TF_FLOAT Double DataType = C.TF_DOUBLE Int32 DataType = C.TF_INT32 + Uint32 DataType = C.TF_UINT32 Uint8 DataType = C.TF_UINT8 Int16 DataType = C.TF_INT16 Int8 DataType = C.TF_INT8 @@ -46,6 +47,7 @@ const ( Complex64 DataType = C.TF_COMPLEX64 Complex DataType = C.TF_COMPLEX Int64 DataType = C.TF_INT64 + Uint64 DataType = C.TF_UINT64 Bool DataType = C.TF_BOOL Qint8 DataType = C.TF_QINT8 Quint8 DataType = C.TF_QUINT8 @@ -217,12 +219,14 @@ var types = []struct { {reflect.TypeOf(float32(0)), C.TF_FLOAT}, {reflect.TypeOf(float64(0)), C.TF_DOUBLE}, {reflect.TypeOf(int32(0)), C.TF_INT32}, + {reflect.TypeOf(uint32(0)), C.TF_UINT32}, {reflect.TypeOf(uint8(0)), C.TF_UINT8}, {reflect.TypeOf(int16(0)), C.TF_INT16}, {reflect.TypeOf(int8(0)), C.TF_INT8}, {reflect.TypeOf(""), C.TF_STRING}, {reflect.TypeOf(complex(float32(0), float32(0))), C.TF_COMPLEX64}, {reflect.TypeOf(int64(0)), C.TF_INT64}, + {reflect.TypeOf(uint64(0)), C.TF_UINT64}, {reflect.TypeOf(false), C.TF_BOOL}, {reflect.TypeOf(uint16(0)), C.TF_UINT16}, {reflect.TypeOf(complex(float64(0), float64(0))), C.TF_COMPLEX128}, |