forked from google-research/google-research
-
Notifications
You must be signed in to change notification settings - Fork 0
/
train_budget.h
64 lines (54 loc) · 2.22 KB
/
train_budget.h
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
// Copyright 2020 The Google Research Authors.
//
// 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.
#ifndef TRAIN_BUDGET_H_
#define TRAIN_BUDGET_H_
#include "generator.h"
#include "instruction.h"
#include "train_budget.pb.h"
namespace automl_zero {
// Class to determine how long a given algorithm should be trained for.
class TrainBudget {
public:
explicit TrainBudget(
// A Algorithm to use as a reference. See `TrainExamples`.
const Algorithm& baseline_algorithm,
// Fraction of the training time of the baseline above which the
// Algorithm will be discarded.
double threshold_factor);
TrainBudget(const TrainBudget& other) = delete;
TrainBudget& operator=(const TrainBudget& other) = delete;
// Returns the number of training examples to use for a given Algorithm.
// In particular, returns 0 if the algorithm should not be evaluated
// altogether (because it is too costly).
// The return value is either `budget` if `algorithm` is less than
// threshold_factor * training time of `baseline_algorithm`
// Otherwise, returns 0, which will cause the algorithm to be assigned
// the minimum fitness.
IntegerT TrainExamples(
// The Algorithm we will be training.
const Algorithm& algorithm,
// The compute budget, measured in training examples.
IntegerT budget) const;
private:
// Cost for running each component function once. Measured in compute-units.
const double baseline_setup_cost_;
const double baseline_train_cost_;
const double threshold_factor_;
};
std::unique_ptr<TrainBudget> BuildTrainBudget(
TrainBudgetSpec train_budget_spec,
// Generator to create baseline.
Generator* generator);
} // namespace automl_zero
#endif // TRAIN_BUDGET_H_