Nigel
/
mindspore
forked from mindspore-Ecosystem/mindspore
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

mindquantum 

mindquantum

cmake lint

projectq patch

test pqc

space

pylint

pqc python ops

doc

check

bug

bug

bug

bug

check boundary condition

mindquantum in log_adapter

mindquantum in adapter and recorder cc

substract to subtract

public variable to private variable

public variable to private variable

commit

debug

debug

const in utils

delete dtol

clint
This commit is contained in:
xuxs 2021-02-04 22:59:00 +08:00 committed by thuxuxs
parent 4b805b87ee
commit c70049b190
39 changed files with 2237 additions and 59 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
cmake/external_libs/projectq.cmake Normal file
View File

@ -0,0 +1,25 @@
set(projectq_CXXFLAGS "-fopenmp -O2 -ffast-mast -march=native -DINTRIN")
set(projectq_CFLAGS "-fopenmp -O2 -ffast-mast -march=native -DINTRIN")
if(ENABLE_GITEE)
set(REQ_URL "https://gitee.com/mirrors/ProjectQ/repository/archive/v0.5.1.tar.gz")
set(MD5 "d874e93e56d3375f1c54c7dd1b731054")
else()
set(REQ_URL "https://github.com/ProjectQ-Framework/ProjectQ/archive/v0.5.1.tar.gz ")
set(MD5 "13430199c253284df8b3d840f11d3560")
endif()
if(ENABLE_CPU AND ${CMAKE_SYSTEM_NAME} MATCHES "Linux"
AND ${CMAKE_HOST_SYSTEM_PROCESSOR} MATCHES "x86_64")
message("Include projectq simulator")
mindspore_add_pkg(projectq
VER 0.5.1
HEAD_ONLY ./
URL ${REQ_URL}
MD5 ${MD5}
PATCHES ${CMAKE_SOURCE_DIR}/third_party/patch/projectq/projectq.patch001
)
include_directories(${projectq_INC})
else()
message("Quantum simulation only support x86_64 linux platform.")
endif()

6
cmake/mind_expression.cmake
View File

@ -48,6 +48,12 @@ if(ENABLE_CPU)
include(${CMAKE_SOURCE_DIR}/cmake/external_libs/mkl_dnn.cmake) include(${CMAKE_SOURCE_DIR}/cmake/external_libs/mkl_dnn.cmake)
endif() endif()
if(ENABLE_CPU AND ${CMAKE_SYSTEM_NAME} MATCHES "Linux"
AND ${CMAKE_HOST_SYSTEM_PROCESSOR} MATCHES "x86_64")
message("Include projectq")
include(${CMAKE_SOURCE_DIR}/cmake/external_libs/projectq.cmake)
endif()
if(ENABLE_GPU) if(ENABLE_GPU)
if(ENABLE_MPI) if(ENABLE_MPI)
include(${CMAKE_SOURCE_DIR}/cmake/external_libs/nccl.cmake) include(${CMAKE_SOURCE_DIR}/cmake/external_libs/nccl.cmake)

11
mindspore/ccsrc/CMakeLists.txt
View File

@ -192,7 +192,7 @@ set(SUB_COMP
frontend/operator frontend/operator
pipeline/jit pipeline/jit
pipeline/pynative pipeline/pynative
common debug pybind_api utils vm profiler ps common debug pybind_api utils vm profiler ps mindquantum
) )
foreach(_comp ${SUB_COMP}) foreach(_comp ${SUB_COMP})
@ -333,13 +333,14 @@ if(CMAKE_SYSTEM_NAME MATCHES "Windows")
target_link_libraries(mindspore mindspore::pybind11_module) target_link_libraries(mindspore mindspore::pybind11_module)
target_link_libraries(mindspore mindspore_gvar) target_link_libraries(mindspore mindspore_gvar)
target_link_libraries(_c_expression PRIVATE -Wl,--whole-archive mindspore -Wl,--no-whole-archive) target_link_libraries(_c_expression PRIVATE -Wl,--whole-archive mindspore -Wl,--no-whole-archive)
elseif (CMAKE_SYSTEM_NAME MATCHES "Darwin") elseif(CMAKE_SYSTEM_NAME MATCHES "Darwin")
target_link_libraries(mindspore mindspore::pybind11_module) target_link_libraries(mindspore mindspore::pybind11_module)
target_link_libraries(mindspore mindspore_gvar) target_link_libraries(mindspore mindspore_gvar)
target_link_libraries(_c_expression PRIVATE -Wl,-force_load mindspore -Wl,-noall_load) target_link_libraries(_c_expression PRIVATE -Wl,-force_load mindspore -Wl,-noall_load)
else () else()
if (ENABLE_CPU AND (ENABLE_D OR ENABLE_GPU)) if(ENABLE_CPU AND (ENABLE_D OR ENABLE_GPU))
target_link_libraries(mindspore mindspore::pslite proto_input mindspore::protobuf mindspore::event mindspore::event_pthreads ${zeromq_DIRPATH}/zmq_install/lib/libzmq.a) target_link_libraries(mindspore mindspore::pslite proto_input mindspore::protobuf
mindspore::event mindspore::event_pthreads ${zeromq_DIRPATH}/zmq_install/lib/libzmq.a)
target_link_libraries(mindspore -Wl,--no-as-needed mindspore::event_core ps_cache) target_link_libraries(mindspore -Wl,--no-as-needed mindspore::event_core ps_cache)
if(${ENABLE_IBVERBS} STREQUAL "ON") if(${ENABLE_IBVERBS} STREQUAL "ON")
target_link_libraries(mindspore ibverbs rdmacm) target_link_libraries(mindspore ibverbs rdmacm)

23
mindspore/ccsrc/backend/kernel_compiler/CMakeLists.txt
View File

@ -37,6 +37,26 @@ if(ENABLE_CPU)
list(REMOVE_ITEM CPU_SRC_LIST "cpu/reduce_scatter_cpu_kernel.cc") list(REMOVE_ITEM CPU_SRC_LIST "cpu/reduce_scatter_cpu_kernel.cc")
list(REMOVE_ITEM CPU_SRC_LIST "cpu/embedding_look_up_comm_grad_cpu_kernel.cc") list(REMOVE_ITEM CPU_SRC_LIST "cpu/embedding_look_up_comm_grad_cpu_kernel.cc")
endif() endif()
file(GLOB_RECURSE QUANTUM_SRC_LIST RELATIVE ${CMAKE_CURRENT_SOURCE_DIR}
"cpu/quantum/*.cc"
)
foreach(qcc ${QUANTUM_SRC_LIST})
list(REMOVE_ITEM CPU_SRC_LIST ${qcc})
endforeach()
if(ENABLE_CPU AND ${CMAKE_SYSTEM_NAME} MATCHES "Linux"
AND ${CMAKE_HOST_SYSTEM_PROCESSOR} MATCHES "x86_64")
message("compiled quantum kernel_compiler")
set_property(SOURCE ${QUANTUM_SRC_LIST} PROPERTY COMPILE_DEFINITIONS
SUBMODULE_ID=mindspore::SubModuleId::SM_MINDQUANTUM)
set_property(SOURCE ${QUANTUM_SRC_LIST} PROPERTY COMPILE_DEFINITIONS INTRIN)
set_property(SOURCE ${QUANTUM_SRC_LIST} PROPERTY COMPILE_OPTIONS -fopenmp -march=native -ffast-math)
else()
message("not compiled quantum kernel_compiler")
set(QUANTUM_SRC_LIST "")
endif()
endif() endif()
if(NOT (ENABLE_CPU AND (ENABLE_D OR ENABLE_GPU))) if(NOT (ENABLE_CPU AND (ENABLE_D OR ENABLE_GPU)))
@ -76,4 +96,5 @@ endif()
set_property(SOURCE ${KERNEL_SRC_LIST} ${CPU_SRC_LIST} ${GPU_SRC_LIST} ${D_SRC_LIST} set_property(SOURCE ${KERNEL_SRC_LIST} ${CPU_SRC_LIST} ${GPU_SRC_LIST} ${D_SRC_LIST}
PROPERTY COMPILE_DEFINITIONS SUBMODULE_ID=mindspore::SubModuleId::SM_KERNEL) PROPERTY COMPILE_DEFINITIONS SUBMODULE_ID=mindspore::SubModuleId::SM_KERNEL)
add_library(_mindspore_backend_kernel_compiler_obj OBJECT ${KERNEL_SRC_LIST} ${CPU_SRC_LIST} ${GPU_SRC_LIST} ${D_SRC_LIST}) add_library(_mindspore_backend_kernel_compiler_obj OBJECT ${KERNEL_SRC_LIST} ${CPU_SRC_LIST}
${GPU_SRC_LIST} ${D_SRC_LIST} ${QUANTUM_SRC_LIST})

248
mindspore/ccsrc/backend/kernel_compiler/cpu/quantum/pqc_cpu_kernel.cc Normal file
View File

@ -0,0 +1,248 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "backend/kernel_compiler/cpu/quantum/pqc_cpu_kernel.h"
#include <omp.h>
#include <utility>
#include <thread>
#include <memory>
#include <algorithm>
#include "utils/ms_utils.h"
#include "runtime/device/cpu/cpu_device_address.h"
#include "common/thread_pool.h"
namespace mindspore {
namespace kernel {
namespace {
struct ComputeParam {
float *encoder_data_cp{nullptr};
float *ansatz_data_cp{nullptr};
float *output_cp{nullptr};
float *gradient_encoder_cp{nullptr};
float *gradient_ansatz_cp{nullptr};
mindquantum::BasicCircuit *circ_cp;
mindquantum::BasicCircuit *herm_circ_cp;
mindquantum::transformer::Hamiltonians *hams_cp;
mindquantum::transformer::NamesType *encoder_params_names_cp;
mindquantum::transformer::NamesType *ansatz_params_names_cp;
std::vector<std::vector<std::shared_ptr<mindquantum::PQCSimulator>>> *tmp_sims_cp;
bool dummy_circuit_cp{false};
size_t result_len_cp{0};
size_t encoder_g_len_cp{0};
size_t ansatz_g_len_cp{0};
};
void ComputerForwardBackward(const std::shared_ptr<ComputeParam> &input_params, size_t start, size_t end, size_t id) {
MS_EXCEPTION_IF_NULL(input_params);
MS_EXCEPTION_IF_NULL(input_params->encoder_data_cp);
MS_EXCEPTION_IF_NULL(input_params->ansatz_data_cp);
MS_EXCEPTION_IF_NULL(input_params->output_cp);
MS_EXCEPTION_IF_NULL(input_params->gradient_encoder_cp);
MS_EXCEPTION_IF_NULL(input_params->gradient_ansatz_cp);
auto encoder_data = input_params->encoder_data_cp;
auto ansatz_data = input_params->ansatz_data_cp;
auto output = input_params->output_cp;
auto gradient_encoder = input_params->gradient_encoder_cp;
auto gradient_ansatz = input_params->gradient_ansatz_cp;
auto circ = input_params->circ_cp;
auto herm_circ = input_params->herm_circ_cp;
auto hams = input_params->hams_cp;
auto encoder_params_names = input_params->encoder_params_names_cp;
auto ansatz_params_names = input_params->ansatz_params_names_cp;
auto tmp_sims = input_params->tmp_sims_cp;
auto dummy_circuit = input_params->dummy_circuit_cp;
auto result_len = input_params->result_len_cp;
auto encoder_g_len = input_params->encoder_g_len_cp;
auto ansatz_g_len = input_params->ansatz_g_len_cp;
MS_EXCEPTION_IF_NULL(hams);
MS_EXCEPTION_IF_NULL(encoder_params_names);
MS_EXCEPTION_IF_NULL(ansatz_params_names);
MS_EXCEPTION_IF_NULL(tmp_sims);
if (end * hams->size() > result_len || end * encoder_params_names->size() * hams->size() > encoder_g_len ||
end * ansatz_params_names->size() * hams->size() > ansatz_g_len) {
MS_LOG(EXCEPTION) << "pqc error input size!";
}
mindquantum::ParameterResolver pr;
for (size_t i = 0; i < ansatz_params_names->size(); i++) {
pr.SetData(ansatz_params_names->at(i), ansatz_data[i]);
}
for (size_t n = start; n < end; ++n) {
for (size_t i = 0; i < encoder_params_names->size(); i++) {
pr.SetData(encoder_params_names->at(i), encoder_data[n * encoder_params_names->size() + i]);
}
auto sim = tmp_sims->at(id)[3];
sim->SetZeroState();
sim->Evolution(*circ, pr);
auto calc_gradient_param = std::make_shared<mindquantum::CalcGradientParam>();
calc_gradient_param->circuit_cp = circ;
calc_gradient_param->circuit_hermitian_cp = herm_circ;
calc_gradient_param->hamiltonians_cp = hams;
calc_gradient_param->paras_cp = &pr;
calc_gradient_param->encoder_params_names_cp = encoder_params_names;
calc_gradient_param->ansatz_params_names_cp = ansatz_params_names;
calc_gradient_param->dummy_circuit_cp = dummy_circuit;
auto e0_grad1_grad_2 =
sim->CalcGradient(calc_gradient_param, *tmp_sims->at(id)[0], *tmp_sims->at(id)[1], *tmp_sims->at(id)[2]);
auto energy = e0_grad1_grad_2[0];
auto grad_encoder = e0_grad1_grad_2[1];
auto grad_ansatz = e0_grad1_grad_2[2];
if (energy.size() != hams->size() || grad_encoder.size() != encoder_params_names->size() * hams->size() ||
grad_ansatz.size() != ansatz_params_names->size() * hams->size()) {
MS_LOG(EXCEPTION) << "pqc error evolution or batch size!";
}
for (size_t poi = 0; poi < hams->size(); poi++) {
output[n * hams->size() + poi] = energy[poi];
}
for (size_t poi = 0; poi < encoder_params_names->size() * hams->size(); poi++) {
gradient_encoder[n * hams->size() * encoder_params_names->size() + poi] = grad_encoder[poi];
}
for (size_t poi = 0; poi < ansatz_params_names->size() * hams->size(); poi++) {
gradient_ansatz[n * hams->size() * ansatz_params_names->size() + poi] = grad_ansatz[poi];
}
}
}
} // namespace
void PQCCPUKernel::InitPQCStructure(const CNodePtr &kernel_node) {
n_threads_user_ = AnfAlgo::GetNodeAttr<int64_t>(kernel_node, mindquantum::kNThreads);
n_qubits_ = AnfAlgo::GetNodeAttr<int64_t>(kernel_node, mindquantum::kNQubits);
encoder_params_names_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::NamesType>(kernel_node, mindquantum::kEncoderParamsNames);
ansatz_params_names_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::NamesType>(kernel_node, mindquantum::kAnsatzParamsNames);
gate_names_ = AnfAlgo::GetNodeAttr<mindquantum::transformer::NamesType>(kernel_node, mindquantum::kGateNames);
gate_matrix_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::ComplexMatrixsType>(kernel_node, mindquantum::kGateMatrix);
gate_obj_qubits_ = AnfAlgo::GetNodeAttr<mindquantum::transformer::Indexess>(kernel_node, mindquantum::kGateObjQubits);
gate_ctrl_qubits_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::Indexess>(kernel_node, mindquantum::kGateCtrlQubits);
gate_params_names_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::ParasNameType>(kernel_node, mindquantum::kGateParamsNames);
gate_coeff_ = AnfAlgo::GetNodeAttr<mindquantum::transformer::CoeffsType>(kernel_node, mindquantum::kGateCoeff);
gate_requires_grad_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::RequiresType>(kernel_node, mindquantum::kGateRequiresGrad);
hams_pauli_coeff_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::PaulisCoeffsType>(kernel_node, mindquantum::kHamsPauliCoeff);
hams_pauli_word_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::PaulisWordsType>(kernel_node, mindquantum::kHamsPauliWord);
hams_pauli_qubit_ =
AnfAlgo::GetNodeAttr<mindquantum::transformer::PaulisQubitsType>(kernel_node, mindquantum::kHamsPauliQubit);
}
void PQCCPUKernel::InitKernel(const CNodePtr &kernel_node) {
MS_EXCEPTION_IF_NULL(kernel_node);
std::vector<size_t> encoder_shape = AnfAlgo::GetInputDeviceShape(kernel_node, 0);
std::vector<size_t> ansatz_shape = AnfAlgo::GetInputDeviceShape(kernel_node, 1);
std::vector<size_t> result_shape = AnfAlgo::GetOutputDeviceShape(kernel_node, 0);
std::vector<size_t> encoder_g_shape = AnfAlgo::GetOutputDeviceShape(kernel_node, 1);
std::vector<size_t> ansatz_g_shape = AnfAlgo::GetOutputDeviceShape(kernel_node, 2);
if (encoder_shape.size() != 2 || ansatz_shape.size() != 1 || result_shape.size() != 2 ||
encoder_g_shape.size() != 3 || ansatz_g_shape.size() != 3) {
MS_LOG(EXCEPTION) << "pqc invalid input size";
}
result_len_ = result_shape[0] * result_shape[1];
encoder_g_len_ = encoder_g_shape[0] * encoder_g_shape[1] * encoder_g_shape[2];
ansatz_g_len_ = ansatz_g_shape[0] * ansatz_g_shape[1] * ansatz_g_shape[2];
n_samples_ = static_cast<unsigned>(encoder_shape[0]);
InitPQCStructure(kernel_node);
dummy_circuit_ = !std::any_of(gate_requires_grad_.begin(), gate_requires_grad_.end(),
[](const mindquantum::transformer::RequireType &rr) {
return std::any_of(rr.begin(), rr.end(), [](const bool &r) { return r; });
});
auto circs = mindquantum::transformer::CircuitTransfor(gate_names_, gate_matrix_, gate_obj_qubits_, gate_ctrl_qubits_,
gate_params_names_, gate_coeff_, gate_requires_grad_);
circ_ = circs[0];
herm_circ_ = circs[1];
hams_ = mindquantum::transformer::HamiltoniansTransfor(hams_pauli_coeff_, hams_pauli_word_, hams_pauli_qubit_);
n_threads_user_ = common::ThreadPool::GetInstance().GetSyncRunThreadNum();
if (n_samples_ < n_threads_user_) {
n_threads_user_ = n_samples_;
}
for (size_t i = 0; i < n_threads_user_; i++) {
tmp_sims_.push_back({});
for (size_t j = 0; j < 4; j++) {
auto tmp = std::make_shared<mindquantum::PQCSimulator>(1, n_qubits_);
tmp_sims_.back().push_back(tmp);
}
}
}
bool PQCCPUKernel::Launch(const std::vector<kernel::AddressPtr> &inputs,
const std::vector<kernel::AddressPtr> & /*workspace*/,
const std::vector<kernel::AddressPtr> &outputs) {
if (inputs.size() != 2 || outputs.size() != 3) {
MS_LOG(EXCEPTION) << "pqc error input output size!";
}
auto encoder_data = reinterpret_cast<float *>(inputs[0]->addr);
auto ansatz_data = reinterpret_cast<float *>(inputs[1]->addr);
auto output = reinterpret_cast<float *>(outputs[0]->addr);
auto gradient_encoder = reinterpret_cast<float *>(outputs[1]->addr);
auto gradient_ansatz = reinterpret_cast<float *>(outputs[2]->addr);
MS_EXCEPTION_IF_NULL(encoder_data);
MS_EXCEPTION_IF_NULL(ansatz_data);
MS_EXCEPTION_IF_NULL(output);
MS_EXCEPTION_IF_NULL(gradient_encoder);
MS_EXCEPTION_IF_NULL(gradient_ansatz);
std::vector<common::Task> tasks;
std::vector<std::shared_ptr<ComputeParam>> thread_params;
tasks.reserve(n_threads_user_);
size_t end = 0;
size_t offset = n_samples_ / n_threads_user_;
size_t left = n_samples_ % n_threads_user_;
for (size_t i = 0; i < n_threads_user_; ++i) {
auto params = std::make_shared<ComputeParam>();
params->encoder_data_cp = encoder_data;
params->ansatz_data_cp = ansatz_data;
params->output_cp = output;
params->gradient_encoder_cp = gradient_encoder;
params->gradient_ansatz_cp = gradient_ansatz;
params->circ_cp = &circ_;
params->herm_circ_cp = &herm_circ_;
params->hams_cp = &hams_;
params->encoder_params_names_cp = &encoder_params_names_;
params->ansatz_params_names_cp = &ansatz_params_names_;
params->tmp_sims_cp = &tmp_sims_;
params->dummy_circuit_cp = dummy_circuit_;
params->result_len_cp = result_len_;
params->encoder_g_len_cp = encoder_g_len_;
params->ansatz_g_len_cp = ansatz_g_len_;
size_t start = end;
end = start + offset;
if (i < left) {
end += 1;
}
auto task = [&params, start, end, i]() {
ComputerForwardBackward(params, start, end, i);
return common::SUCCESS;
};
tasks.emplace_back(task);
thread_params.emplace_back(params);
}
common::ThreadPool::GetInstance().SyncRun(tasks);
return true;
}
} // namespace kernel
} // namespace mindspore

87
mindspore/ccsrc/backend/kernel_compiler/cpu/quantum/pqc_cpu_kernel.h Normal file
View File

@ -0,0 +1,87 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDSPORE_CCSRC_BACKEND_KERNEL_COMPILER_CPU_PQC_CPU_KERNEL_H_
#define MINDSPORE_CCSRC_BACKEND_KERNEL_COMPILER_CPU_PQC_CPU_KERNEL_H_
#include <vector>
#include <memory>
#include <string>
#include "backend/kernel_compiler/cpu/cpu_kernel.h"
#include "backend/kernel_compiler/cpu/cpu_kernel_factory.h"
#include "mindquantum/pqc_simulator.h"
#include "mindquantum/transformer.h"
#include "mindquantum/circuit.h"
#include "mindquantum/parameter_resolver.h"
namespace mindspore {
namespace kernel {
class PQCCPUKernel : public CPUKernel {
public:
PQCCPUKernel() = default;
~PQCCPUKernel() override = default;
void InitKernel(const CNodePtr &kernel_node) override;
bool Launch(const std::vector<AddressPtr> &inputs, const std::vector<AddressPtr> &workspace,
const std::vector<AddressPtr> &outputs) override;
void InitPQCStructure(const CNodePtr &kernel_node);
private:
size_t n_samples_;
size_t n_threads_user_;
bool dummy_circuit_;
size_t result_len_;
size_t encoder_g_len_;
size_t ansatz_g_len_;
int64_t n_qubits_;
mindquantum::BasicCircuit circ_;
mindquantum::BasicCircuit herm_circ_;
mindquantum::transformer::Hamiltonians hams_;
std::vector<std::vector<std::shared_ptr<mindquantum::PQCSimulator>>> tmp_sims_;
// parameters
mindquantum::transformer::NamesType encoder_params_names_;
mindquantum::transformer::NamesType ansatz_params_names_;
// quantum circuit
mindquantum::transformer::NamesType gate_names_;
mindquantum::transformer::ComplexMatrixsType gate_matrix_;
mindquantum::transformer::Indexess gate_obj_qubits_;
mindquantum::transformer::Indexess gate_ctrl_qubits_;
mindquantum::transformer::ParasNameType gate_params_names_;
mindquantum::transformer::CoeffsType gate_coeff_;
mindquantum::transformer::RequiresType gate_requires_grad_;
// hamiltonian
mindquantum::transformer::PaulisCoeffsType hams_pauli_coeff_;
mindquantum::transformer::PaulisWordsType hams_pauli_word_;
mindquantum::transformer::PaulisQubitsType hams_pauli_qubit_;
};
MS_REG_CPU_KERNEL(PQC,
KernelAttr()
.AddInputAttr(kNumberTypeFloat32)
.AddInputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32)
.AddOutputAttr(kNumberTypeFloat32),
PQCCPUKernel);
} // namespace kernel
} // namespace mindspore
#endif // MINDSPORE_CCSRC_BACKEND_KERNEL_COMPILER_CPU_PQC_CPU_KERNEL_H_

51
mindspore/ccsrc/debug/rdr/running_data_recorder.cc
View File

@ -26,31 +26,32 @@ namespace mindspore {
namespace { namespace {
static const char *GetSubModuleName(SubModuleId module_id) { static const char *GetSubModuleName(SubModuleId module_id) {
static const char *sub_module_names[NUM_SUBMODUES] = { static const char *sub_module_names[NUM_SUBMODUES] = {
"UNKNOWN", // SM_UNKNOWN "UNKNOWN", // SM_UNKNOWN
"CORE", // SM_CORE "CORE", // SM_CORE
"ANALYZER", // SM_ANALYZER "ANALYZER", // SM_ANALYZER
"COMMON", // SM_COMMON "COMMON", // SM_COMMON
"DEBUG", // SM_DEBUG "DEBUG", // SM_DEBUG
"DEVICE", // SM_DEVICE "DEVICE", // SM_DEVICE
"GE_ADPT", // SM_GE_ADPT "GE_ADPT", // SM_GE_ADPT
"IR", // SM_IR "IR", // SM_IR
"KERNEL", // SM_KERNEL "KERNEL", // SM_KERNEL
"MD", // SM_MD "MD", // SM_MD
"ME", // SM_ME "ME", // SM_ME
"EXPRESS", // SM_EXPRESS "EXPRESS", // SM_EXPRESS
"OPTIMIZER", // SM_OPTIMIZER "OPTIMIZER", // SM_OPTIMIZER
"PARALLEL", // SM_PARALLEL "PARALLEL", // SM_PARALLEL
"PARSER", // SM_PARSER "PARSER", // SM_PARSER
"PIPELINE", // SM_PIPELINE "PIPELINE", // SM_PIPELINE
"PRE_ACT", // SM_PRE_ACT "PRE_ACT", // SM_PRE_ACT
"PYNATIVE", // SM_PYNATIVE "PYNATIVE", // SM_PYNATIVE
"SESSION", // SM_SESSION "SESSION", // SM_SESSION
"UTILS", // SM_UTILS "UTILS", // SM_UTILS
"VM", // SM_VM "VM", // SM_VM
"PROFILER", // SM_PROFILER "PROFILER", // SM_PROFILER
"PS", // SM_PS "PS", // SM_PS
"LITE", // SM_LITE "LITE", // SM_LITE
"HCCL_ADPT" // SM_HCCL_ADPT "HCCL_ADPT", // SM_HCCL_ADPT
"MINDQUANTUM" // SM_MINDQUANTUM
}; };
return sub_module_names[module_id % NUM_SUBMODUES]; return sub_module_names[module_id % NUM_SUBMODUES];

10
mindspore/ccsrc/mindquantum/CMakeLists.txt Normal file
View File

@ -0,0 +1,10 @@
if(ENABLE_CPU AND ${CMAKE_SYSTEM_NAME} MATCHES "Linux"
AND ${CMAKE_HOST_SYSTEM_PROCESSOR} MATCHES "x86_64")
message("compiled mindquantum")
file(GLOB_RECURSE _MINDQUANTUM_SRC_LIST RELATIVE ${CMAKE_CURRENT_SOURCE_DIR} "*.cc")
set_property(SOURCE ${_MINDQUANTUM_SRC_LIST} PROPERTY COMPILE_DEFINITIONS
SUBMODULE_ID=mindspore::SubModuleId::SM_MINDQUANTUM)
add_library(_mindspore_mindquantum_obj OBJECT ${_MINDQUANTUM_SRC_LIST})
target_compile_options(_mindspore_mindquantum_obj PRIVATE -fopenmp -march=native -ffast-math)
target_compile_definitions(_mindspore_mindquantum_obj PRIVATE INTRIN)
endif()

59
mindspore/ccsrc/mindquantum/circuit.cc Normal file
View File

@ -0,0 +1,59 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/circuit.h"
namespace mindspore {
namespace mindquantum {
BasicCircuit::BasicCircuit() : gate_blocks_({}) {}
void BasicCircuit::AppendBlock() { gate_blocks_.push_back({}); }
void BasicCircuit::AppendNoneParameterGate(const std::string &name, Matrix m, Indexes obj_qubits, Indexes ctrl_qubits) {
auto npg = std::make_shared<NoneParameterGate>(name, m, obj_qubits, ctrl_qubits);
gate_blocks_.back().push_back(npg);
}
void BasicCircuit::AppendParameterGate(const std::string &name, Indexes obj_qubits, Indexes ctrl_qubits,
const ParameterResolver &paras) {
if (name == "RX") {
auto pg_rx = std::make_shared<RXGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_rx);
} else if (name == "RY") {
auto pg_ry = std::make_shared<RYGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_ry);
} else if (name == "RZ") {
auto pg_rz = std::make_shared<RZGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_rz);
} else if (name == "XX") {
auto pg_xx = std::make_shared<XXGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_xx);
} else if (name == "YY") {
auto pg_yy = std::make_shared<YYGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_yy);
} else if (name == "ZZ") {
auto pg_zz = std::make_shared<ZZGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_zz);
} else if (name == "PS") {
auto pg_ps = std::make_shared<PhaseShiftGate>(obj_qubits, ctrl_qubits, paras);
gate_blocks_.back().push_back(pg_ps);
} else {
}
}
const GateBlocks &BasicCircuit::GetGateBlocks() const { return gate_blocks_; }
} // namespace mindquantum
} // namespace mindspore

44
mindspore/ccsrc/mindquantum/circuit.h Normal file
View File

@ -0,0 +1,44 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_CCIRCUIT_H_
#define MINDQUANTUM_ENGINE_CCIRCUIT_H_
#include <vector>
#include <string>
#include <memory>
#include "mindquantum/gates/non_parameter_gate.h"
#include "mindquantum/gates/gates.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
using GateBlock = std::vector<std::shared_ptr<BasicGate>>;
using GateBlocks = std::vector<GateBlock>;
class BasicCircuit {
private:
GateBlocks gate_blocks_;
public:
BasicCircuit();
void AppendBlock();
void AppendNoneParameterGate(const std::string &, Matrix, Indexes, Indexes);
void AppendParameterGate(const std::string &, Indexes, Indexes, const ParameterResolver &);
const GateBlocks &GetGateBlocks() const;
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_CCIRCUIT_H_

46
mindspore/ccsrc/mindquantum/gates/basic_gates.cc Normal file
View File

@ -0,0 +1,46 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/gates/basic_gates.h"
#include <string>
namespace mindspore {
namespace mindquantum {
BasicGate::BasicGate(const std::string &name, bool is_parameter, const Indexes &obj_qubits, const Indexes &ctrl_qubits,
const ParameterResolver &paras)
: name_(name), is_parameter_(is_parameter), obj_qubits_(obj_qubits), ctrl_qubits_(ctrl_qubits), paras_(paras) {}
Matrix BasicGate::GetMatrix(const ParameterResolver &paras_out) {
Matrix gate_matrix_tmp;
return gate_matrix_tmp;
}
Matrix BasicGate::GetDiffMatrix(const ParameterResolver &paras_out) {
Matrix gate_matrix_tmp;
return gate_matrix_tmp;
}
Matrix &BasicGate::GetBaseMatrix() { return gate_matrix_base_; }
const ParameterResolver &BasicGate::GetParameterResolver() const { return paras_; }
bool BasicGate::IsParameterGate() { return is_parameter_; }
Indexes BasicGate::GetObjQubits() { return obj_qubits_; }
Indexes BasicGate::GetCtrlQubits() { return ctrl_qubits_; }
} // namespace mindquantum
} // namespace mindspore

48
mindspore/ccsrc/mindquantum/gates/basic_gates.h Normal file
View File

@ -0,0 +1,48 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_BASIC_GATES_H_
#define MINDQUANTUM_ENGINE_BASIC_GATES_H_
#include <string>
#include "mindquantum/parameter_resolver.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class BasicGate {
private:
std::string name_;
bool is_parameter_;
Matrix gate_matrix_base_;
Indexes obj_qubits_;
Indexes ctrl_qubits_;
ParameterResolver paras_;
public:
BasicGate();
BasicGate(const std::string &, bool, const Indexes &, const Indexes &,
const ParameterResolver &paras = ParameterResolver());
virtual Matrix GetMatrix(const ParameterResolver &);
virtual Matrix GetDiffMatrix(const ParameterResolver &);
virtual Matrix &GetBaseMatrix();
const ParameterResolver &GetParameterResolver() const;
bool IsParameterGate();
Indexes GetObjQubits();
Indexes GetCtrlQubits();
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_BASIC_GATES_H_

154
mindspore/ccsrc/mindquantum/gates/gates.cc Normal file
View File

@ -0,0 +1,154 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/gates/gates.h"
#include <cmath>
namespace mindspore {
namespace mindquantum {
RXGate::RXGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("RX", obj_qubits, ctrl_qubits, paras) {}
RXGate::RXGate() : IntrinsicOneParaGate("RX", {}, {}, {}) {}
Matrix RXGate::GetIntrinsicMatrix(CalcType theta) {
Matrix result = {{{cos(theta / 2), 0}, {0, -sin(theta / 2)}}, {{0, -sin(theta / 2)}, {cos(theta / 2), 0}}};
return result;
}
Matrix RXGate::GetIntrinsicDiffMatrix(CalcType theta) {
Matrix result = {{{-sin(theta / 2) / 2, 0}, {0, -cos(theta / 2) / 2}},
{{0, -cos(theta / 2) / 2}, {-sin(theta / 2) / 2, 0}}};
return result;
}
RYGate::RYGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("RY", obj_qubits, ctrl_qubits, paras) {}
Matrix RYGate::GetIntrinsicMatrix(CalcType theta) {
Matrix result = {{{cos(theta / 2), 0}, {-sin(theta / 2), 0}}, {{sin(theta / 2), 0}, {cos(theta / 2), 0}}};
return result;
}
Matrix RYGate::GetIntrinsicDiffMatrix(CalcType theta) {
Matrix result = {{{-sin(theta / 2) / 2, 0}, {-cos(theta / 2) / 2, 0}},
{{cos(theta / 2) / 2, 0}, {-sin(theta / 2) / 2, 0}}};
return result;
}
RZGate::RZGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("RZ", obj_qubits, ctrl_qubits, paras) {}
Matrix RZGate::GetIntrinsicMatrix(CalcType theta) {
Matrix result = {{{cos(theta / 2), -sin(theta / 2)}, {0, 0}}, {{0, 0}, {cos(theta / 2), sin(theta / 2)}}};
return result;
}
Matrix RZGate::GetIntrinsicDiffMatrix(CalcType theta) {
Matrix result = {{{-sin(theta / 2) / 2, -cos(theta / 2) / 2}, {0, 0}},
{{0, 0}, {-sin(theta / 2) / 2, cos(theta / 2) / 2}}};
return result;
}
PhaseShiftGate::PhaseShiftGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("PS", obj_qubits, ctrl_qubits, paras) {}
Matrix PhaseShiftGate::GetIntrinsicMatrix(CalcType theta) {
Matrix result = {{{1, 0}, {0, 0}}, {{0, 0}, {cos(theta), sin(theta)}}};
return result;
}
Matrix PhaseShiftGate::GetIntrinsicDiffMatrix(CalcType theta) {
Matrix result = {{{0, 0}, {0, 0}}, {{0, 0}, {-sin(theta), cos(theta)}}};
return result;
}
XXGate::XXGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("XX", obj_qubits, ctrl_qubits, paras) {}
Matrix XXGate::GetIntrinsicMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{c, 0}, {0, 0}, {0, 0}, {0, -s}},
{{0, 0}, {c, 0}, {0, -s}, {0, 0}},
{{0, 0}, {0, -s}, {c, 0}, {0, 0}},
{{0, -s}, {0, 0}, {0, 0}, {c, 0}}};
return result;
}
Matrix XXGate::GetIntrinsicDiffMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{-s, 0}, {0, 0}, {0, 0}, {0, -c}},
{{0, 0}, {-s, 0}, {0, -c}, {0, 0}},
{{0, 0}, {0, -c}, {-s, 0}, {0, 0}},
{{0, -c}, {0, 0}, {0, 0}, {-s, 0}}};
return result;
}
YYGate::YYGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("YY", obj_qubits, ctrl_qubits, paras) {}
Matrix YYGate::GetIntrinsicMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{c, 0}, {0, 0}, {0, 0}, {0, s}},
{{0, 0}, {c, 0}, {0, -s}, {0, 0}},
{{0, 0}, {0, -s}, {c, 0}, {0, 0}},
{{0, s}, {0, 0}, {0, 0}, {c, 0}}};
return result;
}
Matrix YYGate::GetIntrinsicDiffMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{-s, 0}, {0, 0}, {0, 0}, {0, c}},
{{0, 0}, {-s, 0}, {0, -c}, {0, 0}},
{{0, 0}, {0, -c}, {-s, 0}, {0, 0}},
{{0, c}, {0, 0}, {0, 0}, {-s, 0}}};
return result;
}
ZZGate::ZZGate(const Indexes &obj_qubits, const Indexes &ctrl_qubits, const ParameterResolver &paras)
: IntrinsicOneParaGate("ZZ", obj_qubits, ctrl_qubits, paras) {}
Matrix ZZGate::GetIntrinsicMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{c, -s}, {0, 0}, {0, 0}, {0, 0}},
{{0, 0}, {c, s}, {0, 0}, {0, 0}},
{{0, 0}, {0, 0}, {c, s}, {0, 0}},
{{0, 0}, {0, 0}, {0, 0}, {c, -s}}};
return result;
}
Matrix ZZGate::GetIntrinsicDiffMatrix(CalcType theta) {
double c = cos(theta);
double s = sin(theta);
Matrix result = {{{-s, -c}, {0, 0}, {0, 0}, {0, 0}},
{{0, 0}, {-s, c}, {0, 0}, {0, 0}},
{{0, 0}, {0, 0}, {-s, c}, {0, 0}},
{{0, 0}, {0, 0}, {0, 0}, {-s, -c}}};
return result;
}
} // namespace mindquantum
} // namespace mindspore

83
mindspore/ccsrc/mindquantum/gates/gates.h Normal file
View File

@ -0,0 +1,83 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_GATES_H_
#define MINDQUANTUM_ENGINE_GATES_H_
#include "mindquantum/gates/intrinsic_one_para_gate.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class RXGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
RXGate(const Indexes &, const Indexes &, const ParameterResolver &);
RXGate();
};
class RYGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
RYGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
class RZGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
RZGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
class PhaseShiftGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
PhaseShiftGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
class XXGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
XXGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
class YYGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
YYGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
class ZZGate : public IntrinsicOneParaGate {
Matrix GetIntrinsicMatrix(CalcType) override;
Matrix GetIntrinsicDiffMatrix(CalcType) override;
public:
ZZGate(const Indexes &, const Indexes &, const ParameterResolver &);
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_GATES_H_

53
mindspore/ccsrc/mindquantum/gates/intrinsic_one_para_gate.cc Normal file
View File

@ -0,0 +1,53 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/gates/intrinsic_one_para_gate.h"
#include <string>
namespace mindspore {
namespace mindquantum {
Matrix IntrinsicOneParaGate::GetIntrinsicMatrix(CalcType theta) {
Matrix gate_matrix_tmp;
return gate_matrix_tmp;
}
Matrix IntrinsicOneParaGate::GetIntrinsicDiffMatrix(CalcType theta) {
Matrix gate_matrix_tmp;
return gate_matrix_tmp;
}
IntrinsicOneParaGate::IntrinsicOneParaGate(const std::string &name, const Indexes &obj_qubits,
const Indexes &ctrl_qubits, const ParameterResolver &paras)
: ParameterGate(name, obj_qubits, ctrl_qubits, paras) {}
CalcType IntrinsicOneParaGate::LinearCombination(const ParameterResolver &paras_in,
const ParameterResolver &paras_out) {
CalcType result = 0;
auto &paras_in_data = paras_in.GetData();
auto &paras_out_data = paras_out.GetData();
for (ParaType::const_iterator i = paras_in_data.begin(); i != paras_in_data.end(); ++i) {
result = result + paras_out_data.at(i->first) * (i->second);
}
return result;
}
Matrix IntrinsicOneParaGate::GetMatrix(const ParameterResolver &paras_out) {
return GetIntrinsicMatrix(LinearCombination(GetParameterResolver(), paras_out));
}
Matrix IntrinsicOneParaGate::GetDiffMatrix(const ParameterResolver &paras_out) {
return GetIntrinsicDiffMatrix(LinearCombination(GetParameterResolver(), paras_out));
}
} // namespace mindquantum
} // namespace mindspore

38
mindspore/ccsrc/mindquantum/gates/intrinsic_one_para_gate.h Normal file
View File

@ -0,0 +1,38 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_INTRINSIC_ONE_PARAGATE_H_
#define MINDQUANTUM_ENGINE_INTRINSIC_ONE_PARAGATE_H_
#include <string>
#include "mindquantum/gates/parameter_gate.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class IntrinsicOneParaGate : public ParameterGate {
virtual Matrix GetIntrinsicMatrix(CalcType);
virtual Matrix GetIntrinsicDiffMatrix(CalcType);
public:
IntrinsicOneParaGate();
IntrinsicOneParaGate(const std::string &, const Indexes &, const Indexes &, const ParameterResolver &);
CalcType LinearCombination(const ParameterResolver &, const ParameterResolver &);
Matrix GetMatrix(const ParameterResolver &) override;
Matrix GetDiffMatrix(const ParameterResolver &) override;
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_INTRINSIC_ONE_PARAGATE_H_

28
mindspore/ccsrc/mindquantum/gates/non_parameter_gate.cc Normal file
View File

@ -0,0 +1,28 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/gates/non_parameter_gate.h"
#include <string>
namespace mindspore {
namespace mindquantum {
NoneParameterGate::NoneParameterGate(const std::string &name, const Matrix &gate_matrix, const Indexes &obj_qubits,
const Indexes &ctrl_qubits)
: BasicGate(name, false, obj_qubits, ctrl_qubits), gate_matrix_(gate_matrix) {}
Matrix &NoneParameterGate::GetBaseMatrix() { return gate_matrix_; }
} // namespace mindquantum
} // namespace mindspore

35
mindspore/ccsrc/mindquantum/gates/non_parameter_gate.h Normal file
View File

@ -0,0 +1,35 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_NON_PARAMETER_GATE_H_
#define MINDQUANTUM_ENGINE_NON_PARAMETER_GATE_H_
#include <string>
#include "mindquantum/gates/basic_gates.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class NoneParameterGate : public BasicGate {
private:
Matrix gate_matrix_;
public:
NoneParameterGate(const std::string &, const Matrix &, const Indexes &, const Indexes &);
Matrix &GetBaseMatrix() override;
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_NON_PARAMETER_GATE_H_

26
mindspore/ccsrc/mindquantum/gates/parameter_gate.cc Normal file
View File

@ -0,0 +1,26 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/gates/parameter_gate.h"
#include <string>
namespace mindspore {
namespace mindquantum {
ParameterGate::ParameterGate(const std::string &name, const Indexes &obj_qubits, const Indexes &ctrl_qubits,
const ParameterResolver &paras)
: BasicGate(name, true, obj_qubits, ctrl_qubits, paras) {}
} // namespace mindquantum
} // namespace mindspore

32
mindspore/ccsrc/mindquantum/gates/parameter_gate.h Normal file
View File

@ -0,0 +1,32 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_PARAMETER_GATE_H_
#define MINDQUANTUM_ENGINE_PARAMETER_GATE_H_
#include <string>
#include "mindquantum/gates/basic_gates.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class ParameterGate : public BasicGate {
public:
ParameterGate();
ParameterGate(const std::string &, const Indexes &, const Indexes &, const ParameterResolver &);
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_PARAMETER_GATE_H_

79
mindspore/ccsrc/mindquantum/hamiltonian.cc Normal file
View File

@ -0,0 +1,79 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/hamiltonian.h"
#include <utility>
namespace mindspore {
namespace mindquantum {
Hamiltonian::Hamiltonian() {}
Hamiltonian::Hamiltonian(const sparse::GoodHamilt &ham, Index n) : ham_(ham), n_qubits_(n) {}
sparse::DequeSparseHam Hamiltonian::TransHamiltonianPhaseOne(int n_thread1, const sparse::GoodHamilt &ham, Index n) {
sparse::DequeSparseHam ham_sparse;
ham_sparse.resize(ham.size());
int step = 0;
#pragma omp parallel for schedule(static) num_threads(n_thread1)
for (Index i = 0; i < ham.size(); i++) {
auto &gt = ham.at(i);
if (gt.second[0].first.size() == 0) {
ham_sparse[i] = sparse::IdentitySparse(n) * gt.first.first * gt.second[0].second;
} else {
ham_sparse[i] = sparse::GoodTerm2Sparse(gt, n);
}
if ((++step) % 20 == 0) std::cout << "\r" << step << "\t/" << ham.size() << "\tfinshed" << std::flush;
}
std::cout << "\ncalculate hamiltonian phase1 finished\n";
return ham_sparse;
}
int Hamiltonian::TransHamiltonianPhaseTwo(sparse::DequeSparseHam &ham_sparse, int n_thread2, int n_split) {
int n = ham_sparse.size();
while (n > 1) {
int half = n / 2 + n % 2;
std::cout << "n: " << n << "\t, half: " << half << "\n";
if (n < n_split) {
break;
}
#pragma omp parallel for schedule(static) num_threads(half)
for (int i = half; i < n; i++) {
ham_sparse[i - half] += ham_sparse[i];
}
ham_sparse.erase(ham_sparse.end() - half + n % 2, ham_sparse.end());
n = half;
}
std::cout << "total: " << ham_sparse.size() << " phase2 finished\n";
return n;
}
void Hamiltonian::SparseHamiltonian(int n_thread1, int n_thread2, int n_split) {
ham_sparse_ = Hamiltonian::TransHamiltonianPhaseOne(n_thread1, ham_, n_qubits_);
final_size_ = Hamiltonian::TransHamiltonianPhaseTwo(ham_sparse_, n_thread2, n_split);
}
void Hamiltonian::SetTermsDict(Simulator::TermsDict const &d) {
td_ = d;
Simulator::ComplexTermsDict().swap(ctd_);
for (auto &term : td_) {
ComplexType coeff = {term.second, 0};
ctd_.push_back(std::make_pair(term.first, coeff));
}
}
void Hamiltonian::Sparsed(bool s) { ham_sparsed_ = s; }
const Simulator::ComplexTermsDict &Hamiltonian::GetCTD() const { return ctd_; }
const Simulator::TermsDict &Hamiltonian::GetTD() const { return td_; }
} // namespace mindquantum
} // namespace mindspore

49
mindspore/ccsrc/mindquantum/hamiltonian.h Normal file
View File

@ -0,0 +1,49 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_CHAMILTONIAN_H_
#define MINDQUANTUM_ENGINE_CHAMILTONIAN_H_
#include "projectq/backends/_sim/_cppkernels/simulator.hpp"
#include "mindquantum/gates/basic_gates.h"
#include "mindquantum/sparse.h"
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class Hamiltonian {
private:
sparse::GoodHamilt ham_;
Index n_qubits_;
sparse::DequeSparseHam ham_sparse_;
Simulator::TermsDict td_;
Simulator::ComplexTermsDict ctd_;
int final_size_ = 1;
bool ham_sparsed_ = false;
public:
Hamiltonian();
Hamiltonian(const sparse::GoodHamilt &, Index);
sparse::DequeSparseHam TransHamiltonianPhaseOne(int, const sparse::GoodHamilt &, Index);
int TransHamiltonianPhaseTwo(sparse::DequeSparseHam &, int, int);
void SparseHamiltonian(int, int, int);
void SetTermsDict(Simulator::TermsDict const &);
void Sparsed(bool);
const Simulator::ComplexTermsDict &GetCTD() const;
const Simulator::TermsDict &GetTD() const;
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_CHAMILTONIAN_H_

34
mindspore/ccsrc/mindquantum/parameter_resolver.cc Normal file
View File

@ -0,0 +1,34 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/parameter_resolver.h"
namespace mindspore {
namespace mindquantum {
ParameterResolver::ParameterResolver()
: data_(ParaType()), no_grad_parameters_(ParaSetType()), requires_grad_parameters_(ParaSetType()) {}
ParameterResolver::ParameterResolver(const ParaType &data, const ParaSetType &no_grad_parameters,
const ParaSetType &requires_grad_parameters)
: data_(data), no_grad_parameters_(no_grad_parameters), requires_grad_parameters_(requires_grad_parameters) {}
const ParaType &ParameterResolver::GetData() const { return data_; }
const ParaSetType &ParameterResolver::GetRequiresGradParameters() const { return requires_grad_parameters_; }
void ParameterResolver::SetData(const std::string &name, const CalcType &value) { data_[name] = value; }
void ParameterResolver::InsertNoGrad(const std::string &name) { no_grad_parameters_.insert(name); }
void ParameterResolver::InsertRequiresGrad(const std::string &name) { requires_grad_parameters_.insert(name); }
} // namespace mindquantum
} // namespace mindspore

44
mindspore/ccsrc/mindquantum/parameter_resolver.h Normal file
View File

@ -0,0 +1,44 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_PARAMETER_RESOLVER_H_
#define MINDQUANTUM_ENGINE_PARAMETER_RESOLVER_H_
#include <map>
#include <string>
#include <set>
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
class ParameterResolver {
public:
ParameterResolver();
ParameterResolver(const ParaType &, const ParaSetType &, const ParaSetType &);
const ParaType &GetData() const;
const ParaSetType &GetRequiresGradParameters() const;
void SetData(const std::string &, const CalcType &);
void InsertNoGrad(const std::string &);
void InsertRequiresGrad(const std::string &);
private:
ParaType data_;
ParaSetType no_grad_parameters_;
ParaSetType requires_grad_parameters_;
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_PARAMETER_RESOLVER_H_

192
mindspore/ccsrc/mindquantum/pqc_simulator.cc Normal file
View File

@ -0,0 +1,192 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/pqc_simulator.h"
#include <omp.h>
#include <numeric>
namespace mindspore {
namespace mindquantum {
PQCSimulator::PQCSimulator() : Simulator(1), n_qubits_(1) {
PQCSimulator::AllocateAll();
for (Index i = 0; i < n_qubits_; i++) {
ordering_.push_back(i);
}
len_ = (1UL << n_qubits_);
}
PQCSimulator::PQCSimulator(Index seed = 1, Index N = 1) : Simulator(seed), n_qubits_(N) {
PQCSimulator::AllocateAll();
for (Index i = 0; i < n_qubits_; i++) {
ordering_.push_back(i);
}
len_ = (1UL << n_qubits_);
}
void PQCSimulator::ApplyGate(std::shared_ptr<BasicGate> g, const ParameterResolver &paras, bool diff) {
if (g->IsParameterGate()) {
if (diff) {
PQCSimulator::apply_controlled_gate(g->GetDiffMatrix(paras), g->GetObjQubits(), g->GetCtrlQubits());
} else {
PQCSimulator::apply_controlled_gate(g->GetMatrix(paras), g->GetObjQubits(), g->GetCtrlQubits());
}
} else {
PQCSimulator::apply_controlled_gate(g->GetBaseMatrix(), g->GetObjQubits(), g->GetCtrlQubits());
}
}
void PQCSimulator::ApplyBlock(const GateBlock &b, const mindquantum::ParameterResolver &paras) {
for (auto &g : b) {
PQCSimulator::ApplyGate(g, paras, false);
}
PQCSimulator::run();
}
void PQCSimulator::ApplyBlocks(const GateBlocks &bs, const ParameterResolver &paras) {
for (auto &b : bs) {
PQCSimulator::ApplyBlock(b, paras);
}
}
void PQCSimulator::Evolution(BasicCircuit const &circuit, ParameterResolver const &paras) {
PQCSimulator::ApplyBlocks(circuit.GetGateBlocks(), paras);
}
CalcType PQCSimulator::Measure(Index mask1, Index mask2, bool apply) {
CalcType out = 0;
#pragma omp parallel for reduction(+ : out) schedule(static)
for (unsigned i = 0; i < (1UL << n_qubits_); i++) {
if (((i & mask1) == mask1) && ((i | mask2) == mask2)) {
out = out + std::real(vec_[i]) * std::real(vec_[i]) + std::imag(vec_[i]) * std::imag(vec_[i]);
} else if (apply) {
vec_[i] = 0;
}
}
return out;
}
std::vector<std::vector<float>> PQCSimulator::CalcGradient(const std::shared_ptr<CalcGradientParam> &input_params,
PQCSimulator &s_left, PQCSimulator &s_right,
PQCSimulator &s_right_tmp) {
// Suppose the simulator already evaluate the circuit.
auto circuit = input_params->circuit_cp;
auto circuit_hermitian = input_params->circuit_hermitian_cp;
auto hamiltonians = input_params->hamiltonians_cp;
auto paras = input_params->paras_cp;
auto encoder_params_names = input_params->encoder_params_names_cp;
auto ansatz_params_names = input_params->ansatz_params_names_cp;
auto dummy_circuit_ = input_params->dummy_circuit_cp;
auto &circ_gate_blocks = circuit->GetGateBlocks();
auto &circ_herm_gate_blocks = circuit_hermitian->GetGateBlocks();
std::map<std::string, size_t> poi;
for (size_t i = 0; i < encoder_params_names->size(); i++) {
poi[encoder_params_names->at(i)] = i;
}
for (size_t i = 0; i < ansatz_params_names->size(); i++) {
poi[ansatz_params_names->at(i)] = i + encoder_params_names->size();
}
if (circ_gate_blocks.size() == 0 || circ_herm_gate_blocks.size() == 0) {
MS_LOG(EXCEPTION) << "Empty quantum circuit!";
}
unsigned len = circ_gate_blocks.at(0).size();
std::vector<float> grad(hamiltonians->size() * poi.size(), 0);
std::vector<float> e0(hamiltonians->size(), 0);
// #pragma omp parallel for
for (size_t h_index = 0; h_index < hamiltonians->size(); h_index++) {
auto &hamiltonian = hamiltonians->at(h_index);
s_right.set_wavefunction(vec_, ordering_);
s_left.set_wavefunction(s_right.vec_, ordering_);
s_left.apply_qubit_operator(hamiltonian.GetCTD(), ordering_);
e0[h_index] = static_cast<float>(ComplexInnerProduct(vec_, s_left.vec_, len_).real());
if (dummy_circuit_) {
continue;
}
for (unsigned i = 0; i < len; i++) {
if ((!circ_herm_gate_blocks.at(0)[i]->IsParameterGate()) ||
(circ_herm_gate_blocks.at(0)[i]->GetParameterResolver().GetRequiresGradParameters().size() == 0)) {
s_left.ApplyGate(circ_herm_gate_blocks.at(0)[i], *paras, false);
s_right.ApplyGate(circ_herm_gate_blocks.at(0)[i], *paras, false);
} else {
s_right.ApplyGate(circ_herm_gate_blocks.at(0)[i], *paras, false);
s_right.run();
s_right_tmp.set_wavefunction(s_right.vec_, ordering_);
s_right_tmp.ApplyGate(circ_gate_blocks.at(0)[len - 1 - i], *paras, true);
s_right_tmp.run();
s_left.run();
ComplexType gi = 0;
if (circ_herm_gate_blocks.at(0)[i]->GetCtrlQubits().size() == 0) {
gi = ComplexInnerProduct(s_left.vec_, s_right_tmp.vec_, len_);
} else {
gi = ComplexInnerProductWithControl(s_left.vec_, s_right_tmp.vec_, len_,
GetControlMask(circ_herm_gate_blocks.at(0)[i]->GetCtrlQubits()));
}
for (auto &it : circ_herm_gate_blocks.at(0)[i]->GetParameterResolver().GetRequiresGradParameters()) {
grad[h_index * poi.size() + poi[it]] -= static_cast<float>(
2 * circ_herm_gate_blocks.at(0)[i]->GetParameterResolver().GetData().at(it) * std::real(gi));
}
s_left.ApplyGate(circ_herm_gate_blocks.at(0)[i], *paras, false);
}
}
}
std::vector<float> grad1;
std::vector<float> grad2;
for (size_t i = 0; i < hamiltonians->size(); i++) {
for (size_t j = 0; j < poi.size(); j++) {
if (j < encoder_params_names->size()) {
grad1.push_back(grad[i * poi.size() + j]);
} else {
grad2.push_back(grad[i * poi.size() + j]);
}
}
}
return {e0, grad1, grad2};
}
void PQCSimulator::AllocateAll() {
for (unsigned i = 0; i < n_qubits_; i++) {
Simulator::allocate_qubit(i);
}
}
void PQCSimulator::DeallocateAll() {
for (unsigned i = 0; i < n_qubits_; i++) {
Simulator::deallocate_qubit(i);
}
}
void PQCSimulator::SetState(const StateVector &wavefunction) { Simulator::set_wavefunction(wavefunction, ordering_); }
std::size_t PQCSimulator::GetControlMask(Indexes const &ctrls) {
std::size_t ctrlmask =
std::accumulate(ctrls.begin(), ctrls.end(), 0, [&](Index a, Index b) { return a | (1UL << ordering_[b]); });
return ctrlmask;
}
void PQCSimulator::ApplyHamiltonian(const Hamiltonian &ham) {
Simulator::apply_qubit_operator(ham.GetCTD(), ordering_);
}
CalcType PQCSimulator::GetExpectationValue(const Hamiltonian &ham) {
return Simulator::get_expectation_value(ham.GetTD(), ordering_);
}
void PQCSimulator::SetZeroState() {
#pragma omp parallel for schedule(static)
for (size_t i = 0; i < len_; i++) {
vec_[i] = {0, 0};
}
vec_[0] = {1, 0};
}
} // namespace mindquantum
} // namespace mindspore

69
mindspore/ccsrc/mindquantum/pqc_simulator.h Normal file
View File

@ -0,0 +1,69 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_PQC_SIMULATOR_H_
#define MINDQUANTUM_ENGINE_PQC_SIMULATOR_H_
#include <map>
#include <vector>
#include <string>
#include <memory>
#include "projectq/backends/_sim/_cppkernels/simulator.hpp"
#include "utils/log_adapter.h"
#include "mindquantum/gates/basic_gates.h"
#include "mindquantum/parameter_resolver.h"
#include "mindquantum/circuit.h"
#include "mindquantum/hamiltonian.h"
#include "mindquantum/utils.h"
#include "mindquantum/transformer.h"
namespace mindspore {
namespace mindquantum {
struct CalcGradientParam {
BasicCircuit *circuit_cp;
BasicCircuit *circuit_hermitian_cp;
transformer::Hamiltonians *hamiltonians_cp;
ParameterResolver *paras_cp;
transformer::NamesType *encoder_params_names_cp;
transformer::NamesType *ansatz_params_names_cp;
bool dummy_circuit_cp{false};
};
class PQCSimulator : public Simulator {
private:
Index n_qubits_;
Indexes ordering_;
Index len_;
public:
PQCSimulator();
PQCSimulator(Index seed, Index N);
void ApplyGate(std::shared_ptr<BasicGate>, const ParameterResolver &, bool);
void ApplyBlock(const GateBlock &, const ParameterResolver &);
void ApplyBlocks(const GateBlocks &, const ParameterResolver &);
void Evolution(const BasicCircuit &, const ParameterResolver &);
CalcType Measure(Index, Index, bool);
void ApplyHamiltonian(const Hamiltonian &);
CalcType GetExpectationValue(const Hamiltonian &);
std::vector<std::vector<float>> CalcGradient(const std::shared_ptr<CalcGradientParam> &, PQCSimulator &,
PQCSimulator &, PQCSimulator &);
void AllocateAll();
void DeallocateAll();
void SetState(const StateVector &);
std::size_t GetControlMask(Indexes const &);
void SetZeroState();
};
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_PQC_SIMULATOR_H_

124
mindspore/ccsrc/mindquantum/sparse.cc Normal file
View File

@ -0,0 +1,124 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/sparse.h"
namespace mindspore {
namespace mindquantum {
namespace sparse {
SparseMatrix BasiGateSparse(char g) {
SparseMatrix out(2, 2);
out.reserve(VectorXi::Constant(2, 2));
switch (g) {
case 'X':
case 'x':
out.insert(0, 1) = 1;
out.insert(1, 0) = 1;
break;
case 'Y':
case 'y':
out.insert(0, 1) = {0, -1};
out.insert(1, 0) = {0, 1};
break;
case 'Z':
case 'z':
out.insert(0, 0) = 1;
out.insert(1, 1) = -1;
break;
case '0':
out.insert(0, 0) = 1;
break;
case '1':
out.insert(1, 1) = 1;
break;
default:
out.insert(0, 0) = 1;
out.insert(1, 1) = 1;
break;
}
out.makeCompressed();
return out;
}
SparseMatrix IdentitySparse(int n_qubit) {
if (n_qubit == 0) {
int dim = 1UL << n_qubit;
SparseMatrix out(dim, dim);
out.reserve(VectorXi::Constant(dim, dim));
for (int i = 0; i < dim; i++) {
out.insert(i, i) = 1;
}
out.makeCompressed();
return out;
} else {
SparseMatrix out = BasiGateSparse('I');
for (int i = 1; i < n_qubit; i++) {
out = KroneckerProductSparse(out, BasiGateSparse('I')).eval();
}
return out;
}
}
SparseMatrix PauliTerm2Sparse(const PauliTerm &pt, Index _min, Index _max) {
int poi;
int n = pt.first.size();
SparseMatrix out;
if (pt.first[0].first == _min) {
out = BasiGateSparse(pt.first[0].second) * pt.second;
poi = 1;
} else {
out = BasiGateSparse('I') * pt.second;
poi = 0;
}
for (Index i = _min + 1; i <= _max; i++) {
if (poi == n) {
out = KroneckerProductSparse(IdentitySparse(_max - i + 1), out).eval();
break;
} else {
if (i == pt.first[poi].first) {
out = KroneckerProductSparse(BasiGateSparse(pt.first[poi++].second), out).eval();
} else {
out = KroneckerProductSparse(BasiGateSparse('I'), out).eval();
}
}
}
return out;
}
SparseMatrix GoodTerm2Sparse(const GoodTerm &gt, Index n_qubits) {
SparseMatrix out = PauliTerm2Sparse(gt.second[0], gt.first.second.first, gt.first.second.second);
for (Index i = 1; i < gt.second.size(); i++) {
out += PauliTerm2Sparse(gt.second[i], gt.first.second.first, gt.first.second.second);
}
out.prune({0.0, 0.0});
out *= gt.first.first;
out = KroneckerProductSparse(out, IdentitySparse(gt.first.second.first)).eval();
out = KroneckerProductSparse(IdentitySparse(n_qubits - gt.first.second.second - 1), out).eval();
return out;
}
} // namespace sparse
} // namespace mindquantum
} // namespace mindspore

48
mindspore/ccsrc/mindquantum/sparse.h Normal file
View File

@ -0,0 +1,48 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_SPARSE_H_
#define MINDQUANTUM_ENGINE_SPARSE_H_
#include <Eigen/Dense>
#include <Eigen/Sparse>
#include <unsupported/Eigen/KroneckerProduct>
#include <deque>
#include <complex>
#include <utility>
#include <iostream>
#include <vector>
#include "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
namespace sparse {
using PauliWord = std::pair<Index, char>;
using PauliTerm = std::pair<std::vector<PauliWord>, int>;
using GoodTerm = std::pair<std::pair<CalcType, std::pair<Index, Index>>, std::vector<PauliTerm>>;
using GoodHamilt = std::vector<GoodTerm>;
typedef Eigen::VectorXcd EigenComplexVector;
using Eigen::VectorXi;
typedef Eigen::SparseMatrix<ComplexType, Eigen::RowMajor, int64_t> SparseMatrix;
using DequeSparseHam = std::deque<SparseMatrix>;
using KroneckerProductSparse = Eigen::KroneckerProductSparse<SparseMatrix, SparseMatrix>;
SparseMatrix BasiGateSparse(char);
SparseMatrix IdentitySparse(int);
SparseMatrix PauliTerm2Sparse(const PauliTerm &, Index, Index);
SparseMatrix GoodTerm2Sparse(const GoodTerm &, Index);
} // namespace sparse
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_SPARSE_H_

115
mindspore/ccsrc/mindquantum/transformer.cc Normal file
View File

@ -0,0 +1,115 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/transformer.h"
#include <algorithm>
#include <utility>
namespace mindspore {
namespace mindquantum {
namespace transformer {
Matrix MatrixConverter(const MatrixType &matrix_real, const MatrixType &matrix_imag, bool hermitian) {
Matrix out;
for (Index i = 0; i < matrix_real.size(); i++) {
out.push_back({});
for (Index j = 0; j < matrix_real.size(); j++) {
if (hermitian)
out.back().push_back({stod(matrix_real[j][i]), -stod(matrix_imag[j][i])});
else
out.back().push_back({stod(matrix_real[i][j]), stod(matrix_imag[i][j])});
}
}
return out;
}
ParameterResolver ParameterResolverConverter(const ParaNameType &para_name, const CoeffType &coeff,
const RequireType &require_grad, bool hermitian) {
ParameterResolver pr;
for (Index i = 0; i < para_name.size(); i++) {
auto name = para_name[i];
if (hermitian)
pr.SetData(name, -coeff[i]);
else
pr.SetData(name, coeff[i]);
if (require_grad[i])
pr.InsertRequiresGrad(name);
else
pr.InsertNoGrad(name);
}
return pr;
}
std::vector<BasicCircuit> CircuitTransfor(const NamesType &names, const ComplexMatrixsType &matrixs,
const Indexess &objs_qubits, const Indexess &ctrls_qubits,
const ParasNameType &paras_name, const CoeffsType &coeffs,
const RequiresType &requires_grad) {
BasicCircuit circuit = BasicCircuit();
BasicCircuit herm_circuit = BasicCircuit();
circuit.AppendBlock();
herm_circuit.AppendBlock();
for (Index n = 0; n < names.size(); n++) {
Indexes obj(objs_qubits[n].size());
Indexes ctrl(ctrls_qubits[n].size());
std::transform(objs_qubits[n].begin(), objs_qubits[n].end(), obj.begin(),
[](const int64_t &i) { return (Index)(i); });
std::transform(ctrls_qubits[n].begin(), ctrls_qubits[n].end(), ctrl.begin(),
[](const int64_t &i) { return (Index)(i); });
if (names[n] == "npg")
// non parameterize gate
circuit.AppendNoneParameterGate("npg", MatrixConverter(matrixs[n][0], matrixs[n][1], false), obj, ctrl);
else
circuit.AppendParameterGate(names[n], obj, ctrl,
ParameterResolverConverter(paras_name[n], coeffs[n], requires_grad[n], false));
}
for (Index n = 0; n < names.size(); n++) {
Index tail = names.size() - 1 - n;
Indexes obj(objs_qubits[tail].size());
Indexes ctrl(ctrls_qubits[tail].size());
std::transform(objs_qubits[tail].begin(), objs_qubits[tail].end(), obj.begin(),
[](const int64_t &i) { return (Index)(i); });
std::transform(ctrls_qubits[tail].begin(), ctrls_qubits[tail].end(), ctrl.begin(),
[](const int64_t &i) { return (Index)(i); });
if (names[tail] == "npg")
// non parameterize gate
herm_circuit.AppendNoneParameterGate("npg", MatrixConverter(matrixs[tail][0], matrixs[tail][1], true), obj, ctrl);
else
herm_circuit.AppendParameterGate(
names[tail], obj, ctrl, ParameterResolverConverter(paras_name[tail], coeffs[tail], requires_grad[tail], true));
}
return {circuit, herm_circuit};
}
Hamiltonians HamiltoniansTransfor(const PaulisCoeffsType &paulis_coeffs, const PaulisWordsType &paulis_words,
const PaulisQubitsType &paulis_qubits) {
Hamiltonians hams;
for (Index n = 0; n < paulis_coeffs.size(); n++) {
Hamiltonian ham;
Simulator::TermsDict td;
for (Index i = 0; i < paulis_coeffs[n].size(); i++) {
Simulator::Term term;
for (Index j = 0; j < paulis_words[n][i].size(); j++)
term.push_back(std::make_pair((Index)(paulis_qubits[n][i][j]), paulis_words[n][i][j].at(0)));
td.push_back(std::make_pair(term, paulis_coeffs[n][i]));
}
ham.SetTermsDict(td);
hams.push_back(ham);
}
return hams;
}
} // namespace transformer
} // namespace mindquantum
} // namespace mindspore

65
mindspore/ccsrc/mindquantum/transformer.h Normal file
View File

@ -0,0 +1,65 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_TRANSFORMER_H_
#define MINDQUANTUM_ENGINE_TRANSFORMER_H_
#include <vector>
#include <string>
#include "mindquantum/gates/gates.h"
#include "mindquantum/circuit.h"
#include "mindquantum/utils.h"
#include "mindquantum/parameter_resolver.h"
#include "mindquantum/hamiltonian.h"
namespace mindspore {
namespace mindquantum {
namespace transformer {
using NameType = std::string;
using MatrixColumnType = std::vector<std::string>;
using MatrixType = std::vector<MatrixColumnType>;
using ComplexMatrixType = std::vector<MatrixType>;
using ParaNameType = std::vector<std::string>;
using CoeffType = std::vector<float>;
using RequireType = std::vector<bool>;
using NamesType = std::vector<NameType>;
using ComplexMatrixsType = std::vector<ComplexMatrixType>;
using ParasNameType = std::vector<ParaNameType>;
using CoeffsType = std::vector<CoeffType>;
using RequiresType = std::vector<RequireType>;
using Indexess = std::vector<std::vector<int64_t>>;
using PauliCoeffsType = std::vector<float>;
using PaulisCoeffsType = std::vector<PauliCoeffsType>;
using PauliWordType = std::vector<std::string>;
using PauliWordsType = std::vector<PauliWordType>;
using PaulisWordsType = std::vector<PauliWordsType>;
using PauliQubitType = std::vector<int64_t>;
using PauliQubitsType = std::vector<PauliQubitType>;
using PaulisQubitsType = std::vector<PauliQubitsType>;
using Hamiltonians = std::vector<Hamiltonian>;
Hamiltonians HamiltoniansTransfor(const PaulisCoeffsType &, const PaulisWordsType &, const PaulisQubitsType &);
std::vector<BasicCircuit> CircuitTransfor(const NamesType &, const ComplexMatrixsType &, const Indexess &,
const Indexess &, const ParasNameType &, const CoeffsType &,
const RequiresType &);
Matrix MatrixConverter(const MatrixType &, const MatrixType &, bool);
ParameterResolver ParameterResolverConverter(const ParaNameType &, const CoeffType &, const RequireType &, bool);
} // namespace transformer
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_TRANSFORMER_H_

50
mindspore/ccsrc/mindquantum/utils.cc Normal file
View File

@ -0,0 +1,50 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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 "mindquantum/utils.h"
namespace mindspore {
namespace mindquantum {
ComplexType ComplexInnerProduct(const Simulator::StateVector &v1, const Simulator::StateVector &v2, unsigned len) {
CalcType real_part = 0;
CalcType imag_part = 0;
#pragma omp parallel for reduction(+ : real_part, imag_part)
for (Index i = 0; i < len; i++) {
real_part += v1[i].real() * v2[i].real() + v1[i].imag() * v2[i].imag();
imag_part += v1[i].real() * v2[i].imag() - v1[i].imag() * v2[i].real();
}
ComplexType result = {real_part, imag_part};
return result;
}
ComplexType ComplexInnerProductWithControl(const Simulator::StateVector &v1, const Simulator::StateVector &v2,
Index len, std::size_t ctrlmask) {
CalcType real_part = 0;
CalcType imag_part = 0;
#pragma omp parallel for reduction(+ : real_part, imag_part)
for (std::size_t i = 0; i < len; i++) {
if ((i & ctrlmask) == ctrlmask) {
real_part += v1[i].real() * v2[i].real() + v1[i].imag() * v2[i].imag();
imag_part += v1[i].real() * v2[i].imag() - v1[i].imag() * v2[i].real();
}
}
ComplexType result = {real_part, imag_part};
return result;
}
} // namespace mindquantum
} // namespace mindspore

56
mindspore/ccsrc/mindquantum/utils.h Normal file
View File

@ -0,0 +1,56 @@
/**
* Copyright 2021 Huawei Technologies Co., Ltd
*
* 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.
*/
#ifndef MINDQUANTUM_ENGINE_UTILS_H_
#define MINDQUANTUM_ENGINE_UTILS_H_
#include <string>
#include <complex>
#include <vector>
#include <map>
#include <set>
#include "projectq/backends/_sim/_cppkernels/intrin/alignedallocator.hpp"
#include "projectq/backends/_sim/_cppkernels/simulator.hpp"
namespace mindspore {
namespace mindquantum {
using CalcType = double;
using ComplexType = std::complex<CalcType>;
using ParaType = std::map<std::string, CalcType>;
using ParaSetType = std::set<std::string>;
using Matrix = std::vector<std::vector<ComplexType, aligned_allocator<ComplexType, 64>>>;
using Index = unsigned;
using Indexes = std::vector<Index>;
using ParaMapType = std::map<std::string, CalcType>;
ComplexType ComplexInnerProduct(const Simulator::StateVector &, const Simulator::StateVector &, Index);
ComplexType ComplexInnerProductWithControl(const Simulator::StateVector &, const Simulator::StateVector &, Index,
std::size_t);
const char kNThreads[] = "n_threads";
const char kNQubits[] = "n_qubits";
const char kEncoderParamsNames[] = "encoder_params_names";
const char kAnsatzParamsNames[] = "ansatz_params_names";
const char kGateNames[] = "gate_names";
const char kGateMatrix[] = "gate_matrix";
const char kGateObjQubits[] = "gate_obj_qubits";
const char kGateCtrlQubits[] = "gate_ctrl_qubits";
const char kGateParamsNames[] = "gate_params_names";
const char kGateCoeff[] = "gate_coeff";
const char kGateRequiresGrad[] = "gate_requires_grad";
const char kHamsPauliCoeff[] = "hams_pauli_coeff";
const char kHamsPauliWord[] = "hams_pauli_word";
const char kHamsPauliQubit[] = "hams_pauli_qubit";
} // namespace mindquantum
} // namespace mindspore
#endif // MINDQUANTUM_ENGINE_UTILS_H_

55
mindspore/core/utils/log_adapter.cc
View File

@ -158,31 +158,32 @@ static std::string ExceptionTypeToString(ExceptionType type) {
static const char *GetSubModuleName(SubModuleId module_id) { static const char *GetSubModuleName(SubModuleId module_id) {
static const char *sub_module_names[NUM_SUBMODUES] = { static const char *sub_module_names[NUM_SUBMODUES] = {
"UNKNOWN", // SM_UNKNOWN "UNKNOWN", // SM_UNKNOWN
"CORE", // SM_CORE "CORE", // SM_CORE
"ANALYZER", // SM_ANALYZER "ANALYZER", // SM_ANALYZER
"COMMON", // SM_COMMON "COMMON", // SM_COMMON
"DEBUG", // SM_DEBUG "DEBUG", // SM_DEBUG
"DEVICE", // SM_DEVICE "DEVICE", // SM_DEVICE
"GE_ADPT", // SM_GE_ADPT "GE_ADPT", // SM_GE_ADPT
"IR", // SM_IR "IR", // SM_IR
"KERNEL", // SM_KERNEL "KERNEL", // SM_KERNEL
"MD", // SM_MD "MD", // SM_MD
"ME", // SM_ME "ME", // SM_ME
"EXPRESS", // SM_EXPRESS "EXPRESS", // SM_EXPRESS
"OPTIMIZER", // SM_OPTIMIZER "OPTIMIZER", // SM_OPTIMIZER
"PARALLEL", // SM_PARALLEL "PARALLEL", // SM_PARALLEL
"PARSER", // SM_PARSER "PARSER", // SM_PARSER
"PIPELINE", // SM_PIPELINE "PIPELINE", // SM_PIPELINE
"PRE_ACT", // SM_PRE_ACT "PRE_ACT", // SM_PRE_ACT
"PYNATIVE", // SM_PYNATIVE "PYNATIVE", // SM_PYNATIVE
"SESSION", // SM_SESSION "SESSION", // SM_SESSION
"UTILS", // SM_UTILS "UTILS", // SM_UTILS
"VM", // SM_VM "VM", // SM_VM
"PROFILER", // SM_PROFILER "PROFILER", // SM_PROFILER
"PS", // SM_PS "PS", // SM_PS
"LITE", // SM_LITE "LITE", // SM_LITE
"HCCL_ADPT" // SM_HCCL_ADPT "HCCL_ADPT", // SM_HCCL_ADPT
"MINDQUANTUM" // SM_MINDQUANTUM
}; };
return sub_module_names[module_id % NUM_SUBMODUES]; return sub_module_names[module_id % NUM_SUBMODUES];
@ -425,7 +426,7 @@ class LogConfigParser {
bool ParseLogLevel(const std::string &str_level, MsLogLevel *ptr_level) { bool ParseLogLevel(const std::string &str_level, MsLogLevel *ptr_level) {
if (str_level.size() == 1) { if (str_level.size() == 1) {
int ch = str_level.c_str()[0]; int ch = str_level.c_str()[0];
ch = ch - '0'; // substract ASCII code of '0', which is 48 ch = ch - '0'; // subtract ASCII code of '0', which is 48
if (ch >= DEBUG && ch <= ERROR) { if (ch >= DEBUG && ch <= ERROR) {
if (ptr_level != nullptr) { if (ptr_level != nullptr) {
*ptr_level = static_cast<MsLogLevel>(ch); *ptr_level = static_cast<MsLogLevel>(ch);
@ -444,7 +445,7 @@ static MsLogLevel GetGlobalLogLevel() {
auto str_level = GetEnv("GLOG_v"); auto str_level = GetEnv("GLOG_v");
if (str_level.size() == 1) { if (str_level.size() == 1) {
int ch = str_level.c_str()[0]; int ch = str_level.c_str()[0];
ch = ch - '0'; // substract ASCII code of '0', which is 48 ch = ch - '0'; // subtract ASCII code of '0', which is 48
if (ch >= DEBUG && ch <= ERROR) { if (ch >= DEBUG && ch <= ERROR) {
log_level = ch; log_level = ch;
} }

1
mindspore/core/utils/log_adapter.h
View File

@ -126,6 +126,7 @@ enum SubModuleId : int {
SM_PS, // Parameter Server SM_PS, // Parameter Server
SM_LITE, // LITE SM_LITE, // LITE
SM_HCCL_ADPT, // Hccl Adapter SM_HCCL_ADPT, // Hccl Adapter
SM_MINDQUANTUM, // MindQuantum
NUM_SUBMODUES // number of submodules NUM_SUBMODUES // number of submodules
}; };

20
mindspore/ops/_grad/grad_other_ops.py
View File

@ -16,6 +16,7 @@
"""Generate bprop for other ops""" """Generate bprop for other ops"""
from .. import operations as P from .. import operations as P
from .. import composite as C
from ..composite.multitype_ops.zeros_like_impl import zeros_like from ..composite.multitype_ops.zeros_like_impl import zeros_like
from .grad_base import bprop_getters from .grad_base import bprop_getters
@ -46,3 +47,22 @@ def get_bprop_iou(self):
def bprop(x, y, out, dout): def bprop(x, y, out, dout):
return zeros_like(x), zeros_like(y) return zeros_like(x), zeros_like(y)
return bprop return bprop
@bprop_getters.register(P.PQC)
def bprop_pqc(self):
"""Generate bprop for PQC"""
t = P.Transpose()
mul = P.Mul()
sum_ = P.ReduceSum()
def bprop(encoder_data, ansatz_data, out, dout):
dx = t(out[1], (2, 0, 1))
dx = mul(dout[0], dx)
dx = sum_(dx, 2)
dx = t(dx, (1, 0))
dy = C.tensor_dot(dout[0], out[2], ((0, 1), (0, 1)))
return dx, dy
return bprop

2
mindspore/ops/operations/__init__.py
View File

@ -98,6 +98,7 @@ from .sparse_ops import SparseToDense
from ._embedding_cache_ops import (CacheSwapHashmap, SearchCacheIdx, CacheSwapTable, UpdateCache, MapCacheIdx, from ._embedding_cache_ops import (CacheSwapHashmap, SearchCacheIdx, CacheSwapTable, UpdateCache, MapCacheIdx,
SubAndFilter, SubAndFilter,
MapUniform, DynamicAssign, PadAndShift) MapUniform, DynamicAssign, PadAndShift)
from .quantum_ops import PQC
__all__ = [ __all__ = [
'Unique', 'Unique',
@ -419,6 +420,7 @@ __all__ = [
"MatrixInverse", "MatrixInverse",
"Range", "Range",
"IndexAdd", "IndexAdd",
"PQC",
] ]
__all__.sort() __all__.sort()

87
mindspore/ops/operations/quantum_ops.py Normal file
View File

@ -0,0 +1,87 @@
# Copyright 2021 Huawei Technologies Co., Ltd
#
# 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.
# ============================================================================
"""Operators for quantum computing."""
from ..primitive import PrimitiveWithInfer, prim_attr_register
from ..._checkparam import Validator as validator
from ...common import dtype as mstype
class PQC(PrimitiveWithInfer):
r"""
Evaluate a parameterized quantum circuit and calculate the gradient of each parameters.
Inputs of this operation is generated by MindQuantum framework.
Inputs:
- **n_qubits** (int) - The qubit number of quantum simulator.
- **encoder_params_names** (List[str]) - The parameters names of encoder circuit.
- **ansatz_params_names** (List[str]) - The parameters names of ansatz circuit.
- **gate_names** (List[str]) - The name of each gate.
- **gate_matrix** (List[List[List[List[float]]]]) - Real part and image
part of the matrix of quantum gate.
- **gate_obj_qubits** (List[List[int]]) - Object qubits of each gate.
- **gate_ctrl_qubits** (List[List[int]]) - Control qubits of each gate.
- **gate_params_names** (List[List[str]]) - Parameter names of each gate.
- **gate_coeff** (List[List[float]]) - Coefficient of eqch parameter of each gate.
- **gate_requires_grad** (List[List[bool]]) - Whether to calculate gradient
of parameters of gates.
- **hams_pauli_coeff** (List[List[float]]) - Coefficient of pauli words.
- **hams_pauli_word** (List[List[List[str]]]) - Pauli words.
- **hams_pauli_qubit** (List[List[List[int]]]) - The qubit that pauli matrix act on.
- **n_threads** (int) - Thread to evaluate input data.
Outputs:
- **expected_value** (Tensor) - The expected value of hamiltonian.
- **g1** (Tensor) - Gradient of encode circuit parameters.
- **g2** (Tensor) - Gradient of ansatz circuit parameters.
Supported Platforms:
``CPU``
"""
@prim_attr_register
def __init__(self, n_qubits, encoder_params_names, ansatz_params_names,
gate_names, gate_matrix, gate_obj_qubits, gate_ctrl_qubits,
gate_params_names, gate_coeff, gate_requires_grad,
hams_pauli_coeff, hams_pauli_word, hams_pauli_qubit,
n_threads):
self.init_prim_io_names(
inputs=['encoder_data', 'ansatz_data'],
outputs=['results', 'encoder_gradient', 'ansatz_gradient'])
self.n_hams = len(hams_pauli_coeff)
def check_shape_size(self, encoder_data, ansatz_data):
if len(encoder_data) != 2:
raise ValueError(
"PQC input encoder_data should have dimension size \
equal to 2, but got {}.".format(len(encoder_data)))
if len(ansatz_data) != 1:
raise ValueError(
"PQC input ansatz_data should have dimension size \
equal to 1, but got {}.".format(len(ansatz_data)))
def infer_shape(self, encoder_data, ansatz_data):
self.check_shape_size(encoder_data, ansatz_data)
return [encoder_data[0], self.n_hams], [
encoder_data[0], self.n_hams,
len(self.encoder_params_names)
], [encoder_data[0], self.n_hams,
len(self.ansatz_params_names)]
def infer_dtype(self, encoder_data, ansatz_data):
args = {'encoder_data': encoder_data, 'ansatz_data': ansatz_data}
validator.check_tensors_dtypes_same_and_valid(args, mstype.float_type,
self.name)
return encoder_data, encoder_data, encoder_data

58
tests/ut/python/ops/test_ops.py
View File

@ -2753,7 +2753,63 @@ test_case_quant_ops = [
'skip': ['backward']}), 'skip': ['backward']}),
] ]
test_case_lists = [test_case_nn_ops, test_case_math_ops, test_case_array_ops, test_case_other_ops, test_case_quant_ops] test_case_quantum_ops = [
('PQC', {
'block': P.PQC(n_qubits=3,
encoder_params_names=['e0', 'e1', 'e2'],
ansatz_params_names=['a', 'b', 'c'],
gate_names=['RX', 'RX', 'RX', 'npg', 'npg',
'npg', 'RX', 'npg', 'npg', 'RZ',
'npg', 'npg', 'RY'],
gate_matrix=[[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.7071067811865475', '0.7071067811865475'],
['0.7071067811865475', '-0.7071067811865475']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.7071067811865475', '0.7071067811865475'],
['0.7071067811865475', '-0.7071067811865475']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.7071067811865475', '0.7071067811865475'],
['0.7071067811865475', '-0.7071067811865475']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '1.0'], ['1.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '-0.0'], ['0.0', '0.0']],
[['0.0', '-1.0'], ['1.0', '0.0']]],
[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '1.0'], ['1.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['1.0', '0.0'], ['0.0', '-1.0']],
[['0.0', '0.0'], ['0.0', '0.0']]],
[[['0.0', '0.0'], ['0.0', '0.0']],
[['0.0', '0.0'], ['0.0', '0.0']]]],
gate_obj_qubits=[[0], [1], [2], [0], [1], [2],
[0], [1], [2], [1], [1], [0], [2]],
gate_ctrl_qubits=[[], [], [], [], [], [], [], [], [], [], [2], [], []],
gate_params_names=[['e0'], ['e1'], ['e2'], [], [], [], ['a'], [], [],
['b'], [], [], ['c']],
gate_coeff=[[1.0], [1.0], [1.0], [], [], [], [1.0], [], [], [1.0], [],
[], [1.0]],
gate_requires_grad=[[True], [True], [True], [], [], [], [True], [], [],
[True], [], [], [True]],
hams_pauli_coeff=[[1.0]],
hams_pauli_word=[[['X', 'Y', 'Z']]],
hams_pauli_qubit=[[[0, 1, 2]]],
n_threads=1),
'desc_inputs': [Tensor(np.array([[1.0, 2.0, 3.0]]).astype(np.float32)),
Tensor(np.array([2.0, 3.0, 4.0]).astype(np.float32))],
'skip': ['backward']}),
]
test_case_lists = [test_case_nn_ops, test_case_math_ops, test_case_array_ops,
test_case_other_ops, test_case_quant_ops, test_case_quantum_ops]
test_case = functools.reduce(lambda x, y: x + y, test_case_lists) test_case = functools.reduce(lambda x, y: x + y, test_case_lists)
# use -k to select certain testcast # use -k to select certain testcast
# pytest tests/python/ops/test_ops.py::test_backward -k LayerNorm # pytest tests/python/ops/test_ops.py::test_backward -k LayerNorm

41
third_party/patch/projectq/projectq.patch001 vendored Normal file
View File

@ -0,0 +1,41 @@
--- ProjectQ-0.5.1/projectq/backends/_sim/_cppkernels/simulator.hpp 2020-06-05 21:07:57.000000000 +0800
+++ ProjectQ-0.5.1_new/projectq/backends/_sim/_cppkernels/simulator.hpp 2021-01-14 10:52:24.822039389 +0800
@@ -33,7 +33,6 @@
#include <random>
#include <functional>
-
class Simulator{
public:
using calc_type = double;
@@ -44,8 +43,9 @@ public:
using Term = std::vector<std::pair<unsigned, char>>;
using TermsDict = std::vector<std::pair<Term, calc_type>>;
using ComplexTermsDict = std::vector<std::pair<Term, complex_type>>;
+ StateVector vec_;
- Simulator(unsigned seed = 1) : N_(0), vec_(1,0.), fusion_qubits_min_(4),
+ Simulator(unsigned seed = 1) : vec_(1,0.), N_(0), fusion_qubits_min_(4),
fusion_qubits_max_(5), rnd_eng_(seed) {
vec_[0]=1.; // all-zero initial state
std::uniform_real_distribution<double> dist(0., 1.);
@@ -562,7 +562,6 @@ private:
}
unsigned N_; // #qubits
- StateVector vec_;
Map map_;
Fusion fused_gates_;
unsigned fusion_qubits_min_, fusion_qubits_max_;
@@ -570,10 +569,8 @@ private:
std::function<double()> rng_;
// large array buffers to avoid costly reallocations
- static StateVector tmpBuff1_, tmpBuff2_;
+ StateVector tmpBuff1_, tmpBuff2_;
};
-Simulator::StateVector Simulator::tmpBuff1_;
-Simulator::StateVector Simulator::tmpBuff2_;
#endif