| /external/eigen/unsupported/test/ |
| D | cxx11_tensor_reduction_cuda.cu | 80 Eigen::TensorMap<Eigen::Tensor<Type, 2, DataLayout> > gpu_out(out_data, dim_y, dim_z); in test_first_dim_reductions() local
84 gpu_out.device(dev) = gpu_in.sum(red_axis); in test_first_dim_reductions()
85 gpu_out.device(dev) += gpu_in.sum(red_axis); in test_first_dim_reductions()
87 dev.memcpyDeviceToHost(redux_gpu.data(), out_data, gpu_out.size()*sizeof(Type)); in test_first_dim_reductions()
91 for (int i = 0; i < gpu_out.size(); ++i) { in test_first_dim_reductions()
120 Eigen::TensorMap<Eigen::Tensor<Type, 2, DataLayout> > gpu_out(out_data, dim_x, dim_y); in test_last_dim_reductions() local
124 gpu_out.device(dev) = gpu_in.sum(red_axis); in test_last_dim_reductions()
125 gpu_out.device(dev) += gpu_in.sum(red_axis); in test_last_dim_reductions()
127 dev.memcpyDeviceToHost(redux_gpu.data(), out_data, gpu_out.size()*sizeof(Type)); in test_last_dim_reductions()
131 for (int i = 0; i < gpu_out.size(); ++i) { in test_last_dim_reductions()
|
| D | cxx11_tensor_random_cuda.cu | 36 Eigen::TensorMap<Eigen::Tensor<float, 2> > gpu_out(d_out, 72,97); in test_cuda_random_uniform() local
38 gpu_out.device(gpu_device) = gpu_out.random(); in test_cuda_random_uniform()
61 Eigen::TensorMap<Eigen::Tensor<float, 2> > gpu_out(d_out, 72,97); in test_cuda_random_normal() local
64 gpu_out.device(gpu_device) = gpu_out.random(gen); in test_cuda_random_normal()
|
| D | cxx11_tensor_complex_cwise_ops_cuda.cu | 41 Eigen::TensorMap<Eigen::Tensor<std::complex<T>, 1, 0, int>, Eigen::Aligned> gpu_out( in test_cuda_complex_cwise_ops() local
62 gpu_out.device(gpu_device) = gpu_in1 + gpu_in2; in test_cuda_complex_cwise_ops()
66 gpu_out.device(gpu_device) = gpu_in1 - gpu_in2; in test_cuda_complex_cwise_ops()
70 gpu_out.device(gpu_device) = gpu_in1 * gpu_in2; in test_cuda_complex_cwise_ops()
74 gpu_out.device(gpu_device) = gpu_in1 / gpu_in2; in test_cuda_complex_cwise_ops()
|
| D | cxx11_tensor_sycl.cpp | 52 TensorMap<Tensor<float, 3>> gpu_out(gpu_out_data, tensorRange); in test_sycl_cpu() local
67 gpu_out.device(sycl_device) = gpu_in1 * 1.2f; in test_sycl_cpu()
81 gpu_out.device(sycl_device) = gpu_in1 * gpu_in2; in test_sycl_cpu()
95 gpu_out.device(sycl_device) = gpu_in1 + gpu_in2; in test_sycl_cpu()
109 gpu_out.device(sycl_device) = gpu_in1 * gpu_in1; in test_sycl_cpu()
123 …gpu_out.device(sycl_device) = gpu_in1 * gpu_in1.constant(3.14f) + gpu_in2 * gpu_in2.constant(2.7f… in test_sycl_cpu()
138 gpu_out.device(sycl_device) =(gpu_in1 > gpu_in1.constant(0.5f)).select(gpu_in2, gpu_in3); in test_sycl_cpu()
|
| D | cxx11_tensor_cuda.cu | 96 Eigen::TensorMap<Eigen::Tensor<float, 1>, Eigen::Aligned> gpu_out( in test_cuda_elementwise_small() local
99 gpu_out.device(gpu_device) = gpu_in1 + gpu_in2; in test_cuda_elementwise_small()
150 …Eigen::TensorMap<Eigen::Tensor<float, 3> > gpu_out(d_out, Eigen::array<Eigen::DenseIndex, 3>(72,53… in test_cuda_elementwise() local
152 gpu_out.device(gpu_device) = gpu_in1 + gpu_in2 * gpu_in3; in test_cuda_elementwise()
191 Eigen::TensorMap<Eigen::Tensor<bool, 1>, Eigen::Aligned> gpu_out( in test_cuda_props() local
194 gpu_out.device(gpu_device) = (gpu_in1.isnan)(); in test_cuda_props()
228 Eigen::TensorMap<Eigen::Tensor<float, 2> > gpu_out(d_out, 72,97); in test_cuda_reduction() local
234 gpu_out.device(gpu_device) = gpu_in1.maximum(reduction_axis); in test_cuda_reduction()
346 Eigen::TensorMap<Eigen::Tensor<float, 4, DataLayout> > gpu_out(d_out, 74,34,11,137); in test_cuda_convolution_1d() local
349 gpu_out.device(gpu_device) = gpu_input.convolve(gpu_kernel, dims); in test_cuda_convolution_1d()
[all …]
|
| D | cxx11_tensor_argmax_cuda.cu | 112 Eigen::TensorMap<Eigen::Tensor<DenseIndex, 3, DataLayout>, Aligned > gpu_out(d_out, out_shape); in test_cuda_argmax_dim() local
114 gpu_out.device(gpu_device) = gpu_in.argmax(dim); in test_cuda_argmax_dim()
142 gpu_out.device(gpu_device) = gpu_in.argmax(dim); in test_cuda_argmax_dim()
201 Eigen::TensorMap<Eigen::Tensor<DenseIndex, 3, DataLayout>, Aligned > gpu_out(d_out, out_shape); in test_cuda_argmin_dim() local
203 gpu_out.device(gpu_device) = gpu_in.argmin(dim); in test_cuda_argmin_dim()
231 gpu_out.device(gpu_device) = gpu_in.argmin(dim); in test_cuda_argmin_dim()
|
| D | cxx11_tensor_broadcast_sycl.cpp | 51 TensorMap<Tensor<float, 4>> gpu_out(gpu_out_data, out_range); in test_broadcast_sycl() local
53 gpu_out.device(sycl_device) = gpu_in.broadcast(broadcasts); in test_broadcast_sycl()
|
| D | cxx11_tensor_forced_eval_sycl.cpp | 45 Eigen::TensorMap<Eigen::Tensor<float, 3>> gpu_out(gpu_out_data, tensorRange); in test_forced_eval_sycl() local
49 gpu_out.device(sycl_device) =(gpu_in1 + gpu_in2).eval() * gpu_in2; in test_forced_eval_sycl()
|
| D | cxx11_tensor_device.cu | 295 Eigen::TensorMap<Eigen::Tensor<float, 3> > gpu_out(d_out, 40,50,70); in test_gpu() local
297 GPUContext context(gpu_in1, gpu_in2, gpu_out); in test_gpu()
|