File: predictor_config.cc

package info (click to toggle)
pytorch 1.13.1%2Bdfsg-4
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 139,252 kB
  • sloc: cpp: 1,100,274; python: 706,454; ansic: 83,052; asm: 7,618; java: 3,273; sh: 2,841; javascript: 612; makefile: 323; xml: 269; ruby: 185; yacc: 144; objc: 68; lex: 44
file content (97 lines) | stat: -rw-r--r-- 2,698 bytes parent folder | download | duplicates (2)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
#include "predictor_config.h"

#include <atomic>

#include "caffe2/core/init.h"
#include "caffe2/utils/proto_utils.h"
#ifdef CAFFE2_OPTIMIZER
#include "caffe2/opt/optimizer.h"
#endif

namespace caffe2 {

namespace {

// We don't use the getNet() from predictor_utils.cc here because that file
// has additional dependencies that we want to avoid bringing in, to keep the
// binary size as small as possible.
const NetDef& getNet(const MetaNetDef& def, const std::string& name) {
  for (const auto& n : def.nets()) {
    if (n.key() == name) {
      return n.value();
    }
  }
  CAFFE_THROW("Net not found: ", name);
}

const ::google::protobuf::RepeatedPtrField<::std::string>& getBlobs(
    const MetaNetDef& def,
    const std::string& name) {
  for (const auto& b : def.blobs()) {
    if (b.key() == name) {
      return b.value();
    }
  }
  CAFFE_THROW("Blob not found: ", name);
}

} // namespace

PredictorConfig
makePredictorConfig(const MetaNetDef& def, Workspace* parent, bool run_init) {
  const auto& init_net =
      getNet(def, PredictorConsts::default_instance().global_init_net_type());
  const auto& run_net =
      getNet(def, PredictorConsts::default_instance().predict_net_type());
  auto config = makePredictorConfig(init_net, run_net, parent, run_init);
  const auto& inputs =
      getBlobs(def, PredictorConsts::default_instance().inputs_blob_type());
  for (const auto& input : inputs) {
    config.input_names.emplace_back(input);
  }

  const auto& outputs =
      getBlobs(def, PredictorConsts::default_instance().outputs_blob_type());
  for (const auto& output : outputs) {
    config.output_names.emplace_back(output);
  }
  return config;
}

PredictorConfig makePredictorConfig(
    const NetDef& init_net,
    const NetDef& run_net,
    Workspace* parent,
    bool run_init,
    int optimization) {
  PredictorConfig config;
  config.ws = std::make_shared<Workspace>(parent);
  auto& ws = *config.ws;
  config.predict_net = std::make_shared<NetDef>(run_net);
  if (run_init) {
    CAFFE_ENFORCE(ws.RunNetOnce(init_net));
  }
#ifdef C10_MOBILE
  GlobalInit();
#endif
  if (optimization &&
      !ArgumentHelper::HasArgument(*config.predict_net, "disable_nomnigraph")) {
#ifdef CAFFE2_OPTIMIZER
    try {
      *config.predict_net =
          opt::optimize(*config.predict_net, &ws, optimization);
    } catch (const std::exception& e) {
      LOG(WARNING) << "Optimization pass failed: " << e.what();
    }
#else
    static std::atomic<bool> warningEmitted{};
    // Emit the log only once.
    if (!warningEmitted.exchange(true)) {
      LOG(WARNING) << "Caffe2 is compiled without optimization passes.";
    }
#endif
  }
  return config;
}

} // namespace caffe2