Home
last modified time | relevance | path

Searched refs:GpuDevice (Results 1 – 25 of 341) sorted by relevance

12345678910>>...14

/external/tensorflow/tensorflow/core/kernels/
Dmirror_pad_op_gpu.cu.cc26 using GpuDevice = Eigen::GpuDevice; typedef
29 template struct functor::MirrorPad<GpuDevice, T, int32, 1>; \
30 template struct functor::MirrorPad<GpuDevice, T, int32, 2>; \
31 template struct functor::MirrorPad<GpuDevice, T, int32, 3>; \
32 template struct functor::MirrorPad<GpuDevice, T, int32, 4>; \
33 template struct functor::MirrorPad<GpuDevice, T, int32, 5>; \
34 template struct functor::MirrorPad<GpuDevice, T, int64, 1>; \
35 template struct functor::MirrorPad<GpuDevice, T, int64, 2>; \
36 template struct functor::MirrorPad<GpuDevice, T, int64, 3>; \
37 template struct functor::MirrorPad<GpuDevice, T, int64, 4>; \
[all …]
Dconv_2d_gpu_float.cu.cc32 template struct ShuffleAndReverse<Eigen::GpuDevice, float, 4, int>;
33 template struct ShuffleAndReverse<Eigen::GpuDevice, float, 4,
36 template struct TransformDepth<Eigen::GpuDevice, float, int>;
38 template struct SwapDimension1And2InTensor3<Eigen::GpuDevice, float4>;
39 template struct SwapDimension1And2InTensor3<Eigen::GpuDevice, float2,
42 template struct SwapDimension0And2InTensor3<Eigen::GpuDevice, float4>;
43 template struct SwapDimension0And2InTensor3<Eigen::GpuDevice, float2,
47 template struct TransformFilter<Eigen::GpuDevice, float, int, 4>;
48 template struct ReverseTransformFilter<Eigen::GpuDevice, float, 4>;
49 template struct NHWCToNCHW<Eigen::GpuDevice, float, 4>;
[all …]
Dconv_2d_gpu_half.cu.cc32 template struct ShuffleAndReverse<Eigen::GpuDevice, Eigen::half, 4, int>;
33 template struct ShuffleAndReverse<Eigen::GpuDevice, Eigen::half, 4,
36 template struct TransformDepth<Eigen::GpuDevice, Eigen::half, int>;
38 template struct SwapDimension1And2InTensor3<Eigen::GpuDevice, Eigen::half>;
41 template struct TransformFilter<Eigen::GpuDevice, Eigen::half, int, 4>;
42 template struct ReverseTransformFilter<Eigen::GpuDevice, Eigen::half, 4>;
43 template struct NHWCToNCHW<Eigen::GpuDevice, Eigen::half, 4>;
44 template struct NCHWToNHWC<Eigen::GpuDevice, Eigen::half, 4>;
45 template struct PadInput<Eigen::GpuDevice, Eigen::half, int, 4>;
48 template struct TransformFilter<Eigen::GpuDevice, Eigen::half, int, 5>;
[all …]
Dconv_2d_gpu_double.cu.cc27 template struct SwapDimension1And2InTensor3<Eigen::GpuDevice, double2,
30 template struct SwapDimension0And2InTensor3<Eigen::GpuDevice, double2,
34 template struct TransformFilter<Eigen::GpuDevice, double, int, 4>;
35 template struct ReverseTransformFilter<Eigen::GpuDevice, double, 4>;
36 template struct NHWCToNCHW<Eigen::GpuDevice, double, 4>;
37 template struct NCHWToNHWC<Eigen::GpuDevice, double, 4>;
38 template struct PadInput<Eigen::GpuDevice, double, int, 4>;
41 template struct TransformFilter<Eigen::GpuDevice, double, int, 5>;
42 template struct ReverseTransformFilter<Eigen::GpuDevice, double, 5>;
43 template struct NHWCToNCHW<Eigen::GpuDevice, double, 5>;
[all …]
Dmaxpooling_op_gpu.h42 const Eigen::GpuDevice& d, bool propagate_nans,
52 const Eigen::GpuDevice& d);
60 const Eigen::GpuDevice& d, const bool include_batch_in_index);
70 T* bottom_diff, const Eigen::GpuDevice& d);
78 const Eigen::GpuDevice& d, const bool include_batch_in_index);
89 const T* top_diff, T* bottom_diff, const Eigen::GpuDevice& d);
Dconcat_lib_gpu.h34 const Eigen::GpuDevice& gpu_device,
40 void ConcatGPUImpl(const Eigen::GpuDevice& d,
49 const Eigen::GpuDevice& gpu_device, \
54 const Eigen::GpuDevice& gpu_device, \
59 const Eigen::GpuDevice& d, \
64 const Eigen::GpuDevice& d, \
Ddepthwise_conv_op_gpu_float.cu.cc23 using Eigen::GpuDevice;
25 template struct LaunchDepthwiseConvOp<GpuDevice, float>;
26 template struct LaunchDepthwiseConvBackpropInputOp<GpuDevice, float>;
27 template struct LaunchDepthwiseConvBackpropFilterOp<GpuDevice, float>;
Ddepthwise_conv_op_gpu_double.cu.cc23 using Eigen::GpuDevice;
25 template struct LaunchDepthwiseConvOp<GpuDevice, double>;
26 template struct LaunchDepthwiseConvBackpropInputOp<GpuDevice, double>;
27 template struct LaunchDepthwiseConvBackpropFilterOp<GpuDevice, double>;
Ddepthwise_conv_op_gpu_half.cu.cc23 using Eigen::GpuDevice;
25 template struct LaunchDepthwiseConvOp<GpuDevice, Eigen::half>;
26 template struct LaunchDepthwiseConvBackpropInputOp<GpuDevice, Eigen::half>;
27 template struct LaunchDepthwiseConvBackpropFilterOp<GpuDevice, Eigen::half>;
Dfused_batch_norm_op.cu.cc25 template struct FusedBatchNormFreezeGrad<Eigen::GpuDevice, float, float>;
26 template struct FusedBatchNormFreezeGrad<Eigen::GpuDevice, Eigen::half, float>;
37 void VarianceToInvVariance<T>::operator()(const Eigen::GpuDevice& d, in operator ()()
59 void InvVarianceToVariance<T>::operator()(const Eigen::GpuDevice& d, in operator ()()
69 void SetNanFunctor<T>::operator()(const Eigen::GpuDevice& d, in operator ()()
Dmirror_pad_op.cc150 using GpuDevice = Eigen::GpuDevice; typedef
206 void MirrorPad<GpuDevice, T, Tpaddings, i>::operator()( \
207 const GpuDevice&, typename TTypes<T, i, int32>::Tensor, \
210 extern template struct MirrorPad<GpuDevice, T, Tpaddings, i>;
236 MirrorPadOp<GpuDevice, T, int32>); \
242 MirrorPadOp<GpuDevice, T, int64>);
412 void MirrorPadGrad<GpuDevice, T, Tpaddings, k>::operator()( \
413 const GpuDevice&, typename TTypes<T, k, int32>::Tensor, \
417 extern template struct MirrorPadGrad<GpuDevice, T, Tpaddings, k>;
443 MirrorPadGradOp<GpuDevice, T, int32>); \
[all …]
Dtile_functor_gpu_float.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, float, int32>;
27 template struct Tile<GpuDevice, float, int64>;
Dtile_functor_gpu_double.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, double, int32>;
27 template struct Tile<GpuDevice, double, int64>;
Dtile_functor_gpu_bool.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, bool, int32>;
27 template struct Tile<GpuDevice, bool, int64>;
Dtile_functor_gpu_int64.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, int64, int32>;
27 template struct Tile<GpuDevice, int64, int64>;
Dtile_functor_gpu_complex128.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, complex128, int32>;
27 template struct Tile<GpuDevice, complex128, int64>;
Dtile_functor_gpu_complex64.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, complex64, int32>;
27 template struct Tile<GpuDevice, complex64, int64>;
Dtile_functor_gpu_int16.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, int16, int32>;
27 template struct Tile<GpuDevice, int16, int64>;
Dtile_functor_gpu_int32.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, int32, int32>;
27 template struct Tile<GpuDevice, int32, int64>;
Dtile_functor_gpu_half.cu.cc24 using Eigen::GpuDevice;
26 template struct Tile<GpuDevice, Eigen::half, int32>;
27 template struct Tile<GpuDevice, Eigen::half, int64>;
Dscan_ops_gpu_float.cu.cc24 using Eigen::GpuDevice;
25 template struct functor::Scan<GpuDevice, Eigen::internal::SumReducer<float>,
27 template struct functor::Scan<GpuDevice, Eigen::internal::ProdReducer<float>,
Dscan_ops_gpu_double.cu.cc24 using Eigen::GpuDevice;
25 template struct functor::Scan<GpuDevice, Eigen::internal::SumReducer<double>,
27 template struct functor::Scan<GpuDevice, Eigen::internal::ProdReducer<double>,
Dscan_ops_gpu_half.cu.cc24 using Eigen::GpuDevice;
26 GpuDevice, Eigen::internal::SumReducer<Eigen::half>, Eigen::half>;
28 GpuDevice, Eigen::internal::ProdReducer<Eigen::half>, Eigen::half>;
/external/eigen/bench/tensors/
Dtensor_benchmarks_fp16_gpu.cu14 Eigen::GpuDevice device(&stream); \
15 BenchmarkSuite<Eigen::GpuDevice, Eigen::half> suite(device, N); \
44 Eigen::GpuDevice device(&stream); \
45 BenchmarkSuite<Eigen::GpuDevice, Eigen::half> suite(device, D1, D2, D3); \
63 Eigen::GpuDevice device(&stream); \
64 BenchmarkSuite<Eigen::GpuDevice, Eigen::half> suite(device, N); \
Dtensor_benchmarks_gpu.cu14 Eigen::GpuDevice device(&stream); \
15 BenchmarkSuite<Eigen::GpuDevice, float> suite(device, N); \
44 Eigen::GpuDevice device(&stream); \
45 BenchmarkSuite<Eigen::GpuDevice, float> suite(device, D1, D2, D3); \
63 Eigen::GpuDevice device(&stream); \
64 BenchmarkSuite<Eigen::GpuDevice, float> suite(device, N); \

12345678910>>...14