| /external/tensorflow/tensorflow/core/kernels/ |
| population_count_op_gpu.cu.cc | 32 typedef Eigen::GpuDevice GPUDevice;
68 void PopulationCount<GPUDevice, T>::operator()( \
71 const GPUDevice& d = c->eigen_device<GPUDevice>(); \
|
| cast_op_impl_bfloat.cc | 23 typedef Eigen::GpuDevice GPUDevice;
46 functor::CastFunctor<GPUDevice, float, bfloat16> func;
47 func(ctx->eigen_device<GPUDevice>(), out->flat<float>(),
|
| cast_op_impl_float.cc | 23 typedef Eigen::GpuDevice GPUDevice;
46 CURRY_TYPES3(CAST_CASE, GPUDevice, float);
47 CAST_CASE(GPUDevice, float, bfloat16);
|
| bucketize_op_gpu.cu.cc | 34 typedef Eigen::GpuDevice GPUDevice;
80 struct BucketizeFunctor<GPUDevice, T> {
86 const GPUDevice& d = context->eigen_device<GPUDevice>();
117 template struct functor::BucketizeFunctor<GPUDevice, type>;
|
| cross_op.cc | 37 typedef Eigen::GpuDevice GPUDevice;
96 void Cross<GPUDevice, type>::operator()( \
97 const GPUDevice& d, TTypes<type, 2>::ConstTensor in0_data, \
100 extern template struct Cross<GPUDevice, type>;
107 CrossOp<GPUDevice, type>);
|
| matrix_diag_op.cc | 41 typedef Eigen::GpuDevice GPUDevice;
179 void MatrixDiag<GPUDevice, T>::Compute( \
180 const GPUDevice& d, typename TTypes<T, 2>::ConstTensor input, \
182 extern template struct MatrixDiag<GPUDevice, T>; \
184 void MatrixDiagPart<GPUDevice, T>::Compute( \
185 const GPUDevice& d, typename TTypes<T, 3>::ConstTensor input, \
187 extern template struct MatrixDiagPart<GPUDevice, T>;
200 MatrixDiagOp<GPUDevice, type>); \
203 MatrixDiagPartOp<GPUDevice, type>)
[all...] |
| depthtospace_op_gpu.cu.cc | 29 using GPUDevice = Eigen::GpuDevice;
144 // Specialization of DepthToSpaceOpFunctor for a GPUDevice.
148 struct DepthToSpaceOpFunctor<GPUDevice, T, FORMAT_NHWC> {
149 void operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input,
167 void operator()(const GPUDevice& d, typename TTypes<T, 5>::ConstTensor input,
174 struct DepthToSpaceOpFunctor<GPUDevice, T, FORMAT_NCHW> {
175 void operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input,
221 void operator()(const GPUDevice& d, typename TTypes<T, 5>::ConstTensor input,
229 template struct functor::DepthToSpaceOpFunctor<GPUDevice, float, FORMAT_NCHW>
[all...] |
| spacetodepth_op_gpu.cu.cc | 28 typedef Eigen::GpuDevice GPUDevice;
144 struct SpaceToDepthOpFunctor<GPUDevice, T, FORMAT_NHWC> {
145 void operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input,
163 void operator()(const GPUDevice& d, typename TTypes<T, 5>::ConstTensor input,
170 struct SpaceToDepthOpFunctor<GPUDevice, T, FORMAT_NCHW> {
171 void operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input,
217 void operator()(const GPUDevice& d, typename TTypes<T, 5>::ConstTensor input,
225 template struct functor::SpaceToDepthOpFunctor<GPUDevice, float, FORMAT_NCHW>;
226 template struct functor::SpaceToDepthOpFunctor<GPUDevice, float, FORMAT_NHWC>
[all...] |
| cudnn_pooling_gpu.cc | 27 typedef Eigen::GpuDevice GPUDevice;
53 functor::NHWCToNCHW<GPUDevice, T, 5>()(context->eigen_device<GPUDevice>(),
109 functor::NCHWToNHWC<GPUDevice, T, 5>()(
110 context->eigen_device<GPUDevice>(),
175 functor::NHWCToNCHW<GPUDevice, T, 5>()(context->eigen_device<GPUDevice>(),
180 functor::NHWCToNCHW<GPUDevice, T, 5>()(context->eigen_device<GPUDevice>(),
[all...] |
| cholesky_op.cc | 76 typedef Eigen::GpuDevice GPUDevice;
81 struct MatrixBandPartFunctor<GPUDevice, T> { \
82 void operator()(OpKernelContext* context, const GPUDevice& device, \
87 extern template struct MatrixBandPartFunctor<GPUDevice, T>;
138 functor::MatrixBandPartFunctor<GPUDevice, Scalar> band_part;
139 band_part(context, context->eigen_device<GPUDevice>(),
|
| matrix_set_diag_op.cc | 40 typedef Eigen::GpuDevice GPUDevice;
147 void MatrixSetDiag<GPUDevice, T>::Compute( \
148 OpKernelContext* context, const GPUDevice& d, \
152 extern template struct MatrixSetDiag<GPUDevice, T>;
165 MatrixSetDiagOp<GPUDevice, type>);
177 MatrixSetDiagOp<GPUDevice, type>);
|
| self_adjoint_eig_v2_op_gpu.cc | 38 typedef Eigen::GpuDevice GPUDevice;
107 const GPUDevice& device = context->eigen_device<GPUDevice>();
110 functor::UnaryFunctor<GPUDevice, functor::conj<Scalar>> conj;
119 functor::UnaryFunctor<GPUDevice, functor::conj<Scalar>> conj;
143 functor::CastFunctor<GPUDevice, Scalar, RealScalar> cast;
|
| bincount_op_gpu.cu.cc | 32 typedef Eigen::GpuDevice GPUDevice;
37 struct BincountFunctor<GPUDevice, T> {
106 template struct functor::BincountFunctor<GPUDevice, type>;
|
| scatter_functor_gpu.cu.h | 30 typedef Eigen::GpuDevice GPUDevice;
75 struct ScatterFunctor<GPUDevice, T, Index, op> {
76 Index operator()(OpKernelContext* c, const GPUDevice& d,
|
| matrix_set_diag_op_gpu.cu.cc | 27 typedef Eigen::GpuDevice GPUDevice;
62 struct MatrixSetDiag<GPUDevice, Scalar> {
63 static void Compute(OpKernelContext* context, const GPUDevice& device,
90 #define DEFINE_GPU_SPEC(T) template struct MatrixSetDiag<GPUDevice, T>;
|
| resize_nearest_neighbor_op_gpu.cu.cc | 31 typedef Eigen::GpuDevice GPUDevice;
97 // Partial specialization of ResizeNearestNeighbor functor for a GPUDevice.
99 struct ResizeNearestNeighbor<GPUDevice, T, align_corners> {
100 bool operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input,
124 template struct ResizeNearestNeighbor<GPUDevice, T, false>; \
125 template struct ResizeNearestNeighbor<GPUDevice, T, true>;
131 // Partial specialization of ResizeNearestNeighborGrad functor for a GPUDevice.
133 struct ResizeNearestNeighborGrad<GPUDevice, T, align_corners> {
134 bool operator()(const GPUDevice& d, typename TTypes<T, 4>::ConstTensor input
[all...] |
| segment_reduction_ops_gpu.cu.cc | 28 using GPUDevice = Eigen::GpuDevice;
129 OpKernelContext* ctx, const GPUDevice& d, const Index output_rows,
170 struct UnsortedSegmentFunctor<GPUDevice, T, Index, InitialValueF, ReductionF> {
180 GPUDevice d = ctx->template eigen_device<GPUDevice>();
214 GPUDevice, T, Index, functor::Lowest<T>, functor::MaxOpGpu<T>>; \
216 GPUDevice, T, Index, functor::Highest<T>, functor::MinOpGpu<T>>; \
217 template struct UnsortedSegmentFunctor<GPUDevice, T, Index, functor::One<T>, \
223 GPUDevice, T, Index, functor::Zero<T>, functor::SumOpGpu<T>>
[all...] |
| strided_slice_op_impl.h | 192 void StridedSlice<GPUDevice, T, NDIM>::operator()( \
193 const GPUDevice& d, typename TTypes<T, NDIM>::Tensor output, \
198 extern template struct StridedSlice<GPUDevice, T, NDIM>; \
200 void Slice<GPUDevice, T, NDIM>::operator()( \
201 const GPUDevice& d, typename TTypes<T, NDIM>::Tensor output, \
205 extern template struct Slice<GPUDevice, T, NDIM>; \
207 void StridedSliceGrad<GPUDevice, T, NDIM>::operator()( \
208 const GPUDevice& d, typename TTypes<T, NDIM>::Tensor output, \
213 extern template struct StridedSliceGrad<GPUDevice, T, NDIM>; \
215 void StridedSliceAssign<GPUDevice, T, NDIM>::operator()(
[all...] |
| where_op.cc | 51 typedef Eigen::GpuDevice GPUDevice;
208 Status NumTrue<GPUDevice, T, Tindex>::Compute( \
209 OpKernelContext* ctx, const GPUDevice& d, TTypes<T>::ConstFlat input, \
211 extern template struct NumTrue<GPUDevice, T, Tindex>
225 Status Where<GPUDevice, Dims, T, Tindex>::Compute( \
226 OpKernelContext* ctx, const GPUDevice& d, \
229 extern template struct Where<GPUDevice, Dims, T, Tindex>;
284 const GPUDevice& d = context->eigen_device<GPUDevice>();
[all...] |
| avgpooling_op.h | 33 if (Eigen::internal::is_same<Device, Eigen::GpuDevice>::value) {
49 typedef Eigen::GpuDevice GPUDevice;
75 const GPUDevice& d);
|
| check_numerics_op_gpu.cu.cc | 32 typedef Eigen::GpuDevice GPUDevice;
61 void Run(const GPUDevice &d, const T *data, int size,
|
| cwise_op_equal_to_1.cc | 32 ApproximateEqualOp<GPUDevice, float>);
35 ApproximateEqualOp<GPUDevice, double>);
|
| /external/tensorflow/tensorflow/contrib/image/kernels/ |
| image_ops.cc | 109 typedef Eigen::GpuDevice GPUDevice;
117 void FillProjectiveTransform<GPUDevice, TYPE>::operator()( \
118 const GPUDevice& device, OutputType* output, const InputType& images, \
120 extern template struct FillProjectiveTransform<GPUDevice, TYPE>
134 ImageProjectiveTransform<GPUDevice, TYPE>)
|
| /external/tensorflow/tensorflow/contrib/reduce_slice_ops/kernels/ |
| reduce_slice_ops_gpu.cu.cc | 28 using GPUDevice = Eigen::GpuDevice;
57 struct ReduceSliceFunctor##reduceop<GPUDevice, T, Index> { \
59 virtual void operator()(OpKernelContext *ctx, const GPUDevice &d, \
86 template struct ReduceSliceFunctor##reduceop<GPUDevice, T, int32>; \
87 template struct ReduceSliceFunctor##reduceop<GPUDevice, T, int64>;
|
| /external/tensorflow/tensorflow/contrib/seq2seq/kernels/ |
| beam_search_ops_gpu.cu.cc | 26 typedef Eigen::GpuDevice GPUDevice;
81 struct GatherTree<GPUDevice, T> {
82 void operator()(OpKernelContext* ctx, const GPUDevice& d,
107 #define DEFINE_GPU_SPECS(T) template struct GatherTree<GPUDevice, T>;
|
