1//===- TrainingLogger.cpp - mlgo feature/reward logging -------------------===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8//
9// This file implements logging infrastructure for extracting features and
10// rewards for mlgo policy training.
11//
12//===----------------------------------------------------------------------===//
13#include "llvm/Analysis/TensorSpec.h"
14#include "llvm/Config/config.h"
15
16#include "llvm/ADT/Twine.h"
17#include "llvm/Analysis/Utils/TrainingLogger.h"
18#include "llvm/Support/CommandLine.h"
19#include "llvm/Support/Debug.h"
20#include "llvm/Support/JSON.h"
21#include "llvm/Support/MemoryBuffer.h"
22#include "llvm/Support/Path.h"
23#include "llvm/Support/raw_ostream.h"
24
25#include <cassert>
26#include <numeric>
27
28using namespace llvm;
29
30void Logger::writeHeader(std::optional<TensorSpec> AdviceSpec) {
31 json::OStream JOS(*OS);
32 JOS.object(Contents: [&]() {
33 JOS.attributeArray(Key: "features", Contents: [&]() {
34 for (const auto &TS : FeatureSpecs)
35 TS.toJSON(OS&: JOS);
36 });
37 if (IncludeReward) {
38 JOS.attributeBegin(Key: "score");
39 RewardSpec.toJSON(OS&: JOS);
40 JOS.attributeEnd();
41 }
42 if (AdviceSpec.has_value()) {
43 JOS.attributeBegin(Key: "advice");
44 AdviceSpec->toJSON(OS&: JOS);
45 JOS.attributeEnd();
46 }
47 });
48 *OS << "\n";
49}
50
51void Logger::switchContext(StringRef Name) {
52 CurrentContext = Name.str();
53 json::OStream JOS(*OS);
54 JOS.object(Contents: [&]() { JOS.attribute(Key: "context", Contents: Name); });
55 *OS << "\n";
56}
57
58void Logger::startObservation() {
59 auto I = ObservationIDs.insert(KV: {CurrentContext, 0});
60 size_t NewObservationID = I.second ? 0 : ++I.first->second;
61 json::OStream JOS(*OS);
62 JOS.object(Contents: [&]() {
63 JOS.attribute(Key: "observation", Contents: static_cast<int64_t>(NewObservationID));
64 });
65 *OS << "\n";
66}
67
68void Logger::endObservation() { *OS << "\n"; }
69
70void Logger::logRewardImpl(const char *RawData) {
71 assert(IncludeReward);
72 json::OStream JOS(*OS);
73 JOS.object(Contents: [&]() {
74 JOS.attribute(Key: "outcome", Contents: static_cast<int64_t>(
75 ObservationIDs.find(Key: CurrentContext)->second));
76 });
77 *OS << "\n";
78 writeTensor(Spec: RewardSpec, RawData);
79 *OS << "\n";
80}
81
82Logger::Logger(std::unique_ptr<raw_ostream> OS,
83 const std::vector<TensorSpec> &FeatureSpecs,
84 const TensorSpec &RewardSpec, bool IncludeReward,
85 std::optional<TensorSpec> AdviceSpec)
86 : OS(std::move(OS)), FeatureSpecs(FeatureSpecs), RewardSpec(RewardSpec),
87 IncludeReward(IncludeReward) {
88 writeHeader(AdviceSpec);
89}
90

source code of llvm/lib/Analysis/TrainingLogger.cpp