analyzer_resnet50_tester.cc 3.3 KB
Newer Older
T
Tao Luo 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
/* Copyright (c) 2018 PaddlePaddle Authors. All Rights Reserved.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License. */

#include <fstream>
#include <iostream>
#include "paddle/fluid/inference/tests/api/tester_helper.h"

namespace paddle {
namespace inference {
namespace analysis {

T
Tao Luo 已提交
23
void SetConfig(AnalysisConfig *cfg) {
T
Tao Luo 已提交
24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
  cfg->param_file = FLAGS_infer_model + "/params";
  cfg->prog_file = FLAGS_infer_model + "/model";
  cfg->use_gpu = false;
  cfg->device = 0;
  cfg->enable_ir_optim = true;
  cfg->specify_input_name = true;
}

void SetInput(std::vector<std::vector<PaddleTensor>> *inputs) {
  PADDLE_ENFORCE_EQ(FLAGS_test_all_data, 0, "Only have single batch of data.");

  PaddleTensor input;
  // channel=3, height/width=318
  std::vector<int> shape({FLAGS_batch_size, 3, 318, 318});
  input.shape = shape;
  input.dtype = PaddleDType::FLOAT32;

  // fill input data, for profile easily, do not use random data here.
  size_t size = FLAGS_batch_size * 3 * 318 * 318;
  input.data.Resize(size * sizeof(float));
  float *input_data = static_cast<float *>(input.data.data());
  for (size_t i = 0; i < size; i++) {
    *(input_data + i) = static_cast<float>(i) / size;
  }

  std::vector<PaddleTensor> input_slots;
  input_slots.assign({input});
  (*inputs).emplace_back(input_slots);
}

// Easy for profiling independently.
T
Tao Luo 已提交
55
void profile(bool use_mkldnn = false) {
T
Tao Luo 已提交
56 57
  AnalysisConfig cfg;
  SetConfig(&cfg);
T
Tao Luo 已提交
58
  cfg._use_mkldnn = use_mkldnn;
T
Tao Luo 已提交
59 60 61 62 63 64 65 66 67 68 69 70 71 72
  std::vector<PaddleTensor> outputs;

  std::vector<std::vector<PaddleTensor>> input_slots_all;
  SetInput(&input_slots_all);
  TestPrediction(cfg, input_slots_all, &outputs, FLAGS_num_threads);

  if (FLAGS_num_threads == 1 && !FLAGS_test_all_data) {
    PADDLE_ENFORCE_EQ(outputs.size(), 1UL);
    size_t size = GetSize(outputs[0]);
    // output is a 512-dimension feature
    EXPECT_EQ(size, 512 * FLAGS_batch_size);
  }
}

T
Tao Luo 已提交
73 74 75 76 77
TEST(Analyzer_resnet50, profile) { profile(); }
#ifndef PADDLE_WITH_MKLDNN
TEST(Analyzer_resnet50, profile_mkldnn) { profile(true /* use_mkldnn */); }
#endif

T
Tao Luo 已提交
78 79 80 81 82
// Check the fuse status
TEST(Analyzer_resnet50, fuse_statis) {
  AnalysisConfig cfg;
  SetConfig(&cfg);
  int num_ops;
T
Tao Luo 已提交
83 84 85 86 87
  auto predictor = CreatePaddlePredictor<AnalysisConfig>(cfg);
  auto fuse_statis = GetFuseStatis(
      static_cast<AnalysisPredictor *>(predictor.get()), &num_ops);
  ASSERT_TRUE(fuse_statis.count("fc_fuse"));
  EXPECT_EQ(fuse_statis.at("fc_fuse"), 1);
T
Tao Luo 已提交
88 89 90
}

// Compare result of NativeConfig and AnalysisConfig
T
Tao Luo 已提交
91
void compare(bool use_mkldnn = false) {
T
Tao Luo 已提交
92 93
  AnalysisConfig cfg;
  SetConfig(&cfg);
T
Tao Luo 已提交
94
  cfg._use_mkldnn = use_mkldnn;
T
Tao Luo 已提交
95 96 97 98

  std::vector<std::vector<PaddleTensor>> input_slots_all;
  SetInput(&input_slots_all);
  CompareNativeAndAnalysis(cfg, input_slots_all);
T
Tao Luo 已提交
99 100
}

T
Tao Luo 已提交
101
TEST(Analyzer_resnet50, compare) { compare(); }
T
Tao Luo 已提交
102
#ifdef PADDLE_WITH_MKLDNN
T
Tao Luo 已提交
103
TEST(Analyzer_resnet50, compare_mkldnn) { compare(true /* use_mkldnn */); }
T
Tao Luo 已提交
104
#endif
T
Tao Luo 已提交
105 106 107 108

}  // namespace analysis
}  // namespace inference
}  // namespace paddle