// Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved. // // Licensed under the Apache License, Version 2.0 (the "License"); // you may not use this file except in compliance with the License. // You may obtain a copy of the License at // // http://www.apache.org/licenses/LICENSE-2.0 // // Unless required by applicable law or agreed to in writing, software // distributed under the License is distributed on an "AS IS" BASIS, // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. // See the License for the specific language governing permissions and // limitations under the License. #include "paddle/pten/backends/gpu/gpu_context.h" #include "paddle/pten/core/kernel_registry.h" #include "paddle/pten/kernels/gpu/reduce.h" #include "paddle/pten/kernels/impl/trace_kernel_impl.h" #include "paddle/pten/kernels/trace_kernel.h" namespace pten { template void TraceKernel(const Context& ctx, const DenseTensor& x, int offset, int axis1, int axis2, DenseTensor* out) { T* out_data = out->mutable_data(ctx.GetPlace()); auto diag = Diagonal(ctx, &x, offset, axis1, axis2); if (diag.numel() > 0) { auto stream = ctx.stream(); std::vector reduce_dims; reduce_dims.push_back(out->dims().size()); kernels:: TensorReduceFunctorImpl>( ctx, diag, out, kps::IdentityFunctor(), reduce_dims, stream); } else { paddle::operators::math::SetConstant functor; functor(ctx, out, static_cast(0)); } } } // namespace pten PT_REGISTER_KERNEL(trace, GPU, ALL_LAYOUT, pten::TraceKernel, float, double, int, int64_t, paddle::platform::float16, paddle::platform::complex, paddle::platform::complex) {}