Program Listing for File paddle_mkldnn_quantizer_config.h¶
↰ Return to documentation for file (docs/paddle_include_file/paddle_mkldnn_quantizer_config.h
)
// Copyright (c) 2019 PaddlePaddle Authors. All Rights Reserved.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#pragma once
#include <cassert>
#include <map>
#include <memory>
#include <string>
#include <unordered_set>
#include <vector>
#include "paddle_api.h" // NOLINT
namespace paddle {
enum class ScaleAlgo {
NONE,
MAX,
MAX_CH,
MAX_CH_T,
KL,
};
struct MkldnnQuantizerConfig {
MkldnnQuantizerConfig();
void SetScaleAlgo(std::string op_type_name, std::string conn_name,
ScaleAlgo algo) {
rules_[op_type_name][conn_name] = algo;
}
ScaleAlgo scale_algo(const std::string& op_type_name,
const std::string& conn_name) const;
void SetWarmupData(std::shared_ptr<std::vector<PaddleTensor>> data) {
warmup_data_ = data;
}
std::shared_ptr<std::vector<PaddleTensor>> warmup_data() const {
return warmup_data_;
}
void SetWarmupBatchSize(int batch_size) { warmup_bs_ = batch_size; }
int warmup_batch_size() const { return warmup_bs_; }
void SetEnabledOpTypes(std::unordered_set<std::string> op_list) {
enabled_op_types_ = op_list;
}
const std::unordered_set<std::string>& enabled_op_types() const {
return enabled_op_types_;
}
void SetExcludedOpIds(std::unordered_set<int> op_ids_list) {
excluded_op_ids_ = op_ids_list;
}
const std::unordered_set<int>& excluded_op_ids() const {
return excluded_op_ids_;
}
void SetDefaultScaleAlgo(ScaleAlgo algo) { default_scale_algo_ = algo; }
ScaleAlgo default_scale_algo() const { return default_scale_algo_; }
protected:
std::map<std::string, std::map<std::string, ScaleAlgo>> rules_;
std::unordered_set<std::string> enabled_op_types_;
std::unordered_set<int> excluded_op_ids_;
std::shared_ptr<std::vector<PaddleTensor>> warmup_data_;
int warmup_bs_{1};
ScaleAlgo default_scale_algo_{ScaleAlgo::MAX};
};
} // namespace paddle