mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-07 00:21:07 +01:00
a296fa36ac
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/52903 Implement BlackBoxPredictor::BenchmarkIndividualOps so that we can clean up the output tensors properly after each iteration and get more accurate per operator timing. Add four more metrics to track setup_time, memory_alloc_time, memory_dealloc_time, and output_dealloc_time. Reviewed By: ajyu Differential Revision: D26657473 fbshipit-source-id: 1cf282192b531513b9ee40b37252087818412f81
97 lines
2.5 KiB
C++
97 lines
2.5 KiB
C++
#ifndef CAFFE2_CORE_NET_SIMPLE_H_
|
|
#define CAFFE2_CORE_NET_SIMPLE_H_
|
|
|
|
#include <vector>
|
|
|
|
#include "c10/util/Registry.h"
|
|
#include "caffe2/core/common.h"
|
|
#include "caffe2/core/logging.h"
|
|
#include "caffe2/core/net.h"
|
|
#include "caffe2/core/tensor.h"
|
|
#include "caffe2/core/workspace.h"
|
|
#include "caffe2/proto/caffe2_pb.h"
|
|
|
|
namespace caffe2 {
|
|
|
|
struct IndividualMetrics {
|
|
public:
|
|
explicit IndividualMetrics(const std::vector<OperatorBase*>& operators)
|
|
: main_runs_(0), operators_(operators) {
|
|
const auto num_ops = operators_.size();
|
|
time_per_op.resize(num_ops, 0.0);
|
|
}
|
|
// run ops while collecting profiling results
|
|
void RunOpsWithProfiling();
|
|
|
|
// print out profiling results
|
|
void PrintOperatorProfilingResults();
|
|
|
|
const vector<float>& GetTimePerOp() {
|
|
return time_per_op;
|
|
}
|
|
|
|
float setup_time{0.0};
|
|
float memory_alloc_time{0.0};
|
|
float memory_dealloc_time{0.0};
|
|
float output_dealloc_time{0.0};
|
|
|
|
private:
|
|
int main_runs_;
|
|
const std::vector<OperatorBase*>& operators_;
|
|
|
|
vector<float> time_per_op;
|
|
vector<uint64_t> flops_per_op;
|
|
vector<uint64_t> memory_bytes_read_per_op;
|
|
vector<uint64_t> memory_bytes_written_per_op;
|
|
vector<uint64_t> param_bytes_per_op;
|
|
|
|
CaffeMap<string, int> num_ops_per_op_type_;
|
|
CaffeMap<string, float> time_per_op_type;
|
|
CaffeMap<string, float> flops_per_op_type;
|
|
CaffeMap<string, float> memory_bytes_read_per_op_type;
|
|
CaffeMap<string, float> memory_bytes_written_per_op_type;
|
|
CaffeMap<string, float> param_bytes_per_op_type;
|
|
};
|
|
|
|
// This is the very basic structure you need to run a network - all it
|
|
// does is simply to run everything in sequence. If you want more fancy control
|
|
// such as a DAG-like execution, check out other better net implementations.
|
|
class TORCH_API SimpleNet : public NetBase {
|
|
public:
|
|
SimpleNet(const std::shared_ptr<const NetDef>& net_def, Workspace* ws);
|
|
bool SupportsAsync() override {
|
|
return false;
|
|
}
|
|
|
|
vector<float> TEST_Benchmark(
|
|
const int warmup_runs,
|
|
const int main_runs,
|
|
const bool run_individual) override;
|
|
|
|
/*
|
|
* This returns a list of pointers to objects stored in unique_ptrs.
|
|
* Used by Observers.
|
|
*
|
|
* Think carefully before using.
|
|
*/
|
|
vector<OperatorBase*> GetOperators() const override {
|
|
vector<OperatorBase*> op_list;
|
|
for (auto& op : operators_) {
|
|
op_list.push_back(op.get());
|
|
}
|
|
return op_list;
|
|
}
|
|
|
|
protected:
|
|
bool Run() override;
|
|
bool RunAsync() override;
|
|
|
|
vector<unique_ptr<OperatorBase>> operators_;
|
|
|
|
C10_DISABLE_COPY_AND_ASSIGN(SimpleNet);
|
|
};
|
|
|
|
} // namespace caffe2
|
|
|
|
#endif // CAFFE2_CORE_NET_SIMPLE_H_
|