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
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
|
#include "distribution.h"
#include <random>
#include <climits>
#include "json.h"
#include "rng.h"
struct int_distribution_impl {
virtual ~int_distribution_impl() = default;
virtual int minimum() const = 0;
virtual int sample() = 0;
virtual std::string description() const = 0;
int blo = 0;
int bhi = INT_MAX;
};
struct fixed_distribution : int_distribution_impl {
int value;
explicit fixed_distribution( int v )
: value( v )
{}
int minimum() const override {
return value;
}
int sample() override {
return value;
}
std::string description() const override {
// NOLINTNEXTLINE(cata-translate-string-literal)
return string_format( "Fixed(%d)", value );
}
};
struct uniform_distribution : int_distribution_impl {
std::uniform_int_distribution<int> dist;
explicit uniform_distribution( int a, int b )
: dist( a, b )
{}
int minimum() const override {
return 0;
}
int sample() override {
return dist( rng_get_engine() );
}
std::string description() const override {
// NOLINTNEXTLINE(cata-translate-string-literal)
return string_format( "Uniform(%d,%d)", dist.a(), dist.b() );
}
};
struct binomial_distribution : int_distribution_impl {
std::binomial_distribution<int> dist;
explicit binomial_distribution( int lo, int hi, int t, double p )
: dist( t, p ) {
blo = lo;
bhi = hi;
}
int minimum() const override {
return 0;
}
int sample() override {
int distvalue = dist( rng_get_engine() );
int rvalue = std::min( distvalue, bhi );
rvalue = std::max( rvalue, blo );
return rvalue;
}
std::string description() const override {
// NOLINTNEXTLINE(cata-translate-string-literal)
return string_format( "Binomial(%d,%.0f)", dist.t(), dist.p() );
}
};
struct poisson_distribution : int_distribution_impl {
std::poisson_distribution<int> dist;
explicit poisson_distribution( int lo, int hi, double mean )
: dist( mean ) {
blo = lo;
bhi = hi;
}
int minimum() const override {
return 0;
}
int sample() override {
int rvalue = std::min( dist( rng_get_engine() ), bhi );
rvalue = std::max( rvalue, blo );
return rvalue;
}
std::string description() const override {
// NOLINTNEXTLINE(cata-translate-string-literal)
return string_format( "Poisson(%.0f)", dist.mean() );
}
};
int_distribution::int_distribution()
: impl_( make_shared_fast<fixed_distribution>( 0 ) )
{}
int_distribution::int_distribution( int v )
: impl_( make_shared_fast<fixed_distribution>( v ) )
{}
int int_distribution::minimum() const
{
return impl_->minimum();
}
int int_distribution::sample() const
{
return impl_->sample();
}
std::string int_distribution::description() const
{
return impl_->description();
}
void int_distribution::deserialize( const JsonValue &jin )
{
int lo = 0;
int hi = INT_MAX;
if( jin.test_int() ) {
int v = jin.get_int();
impl_ = make_shared_fast<fixed_distribution>( v );
} else if( jin.test_array() ) {
JsonArray jarr = jin.get_array();
if( jarr.size() != 2 ) {
jarr.throw_error( "uniform array has wrong number of elements, should be [ lo, hi ]" );
}
lo = jarr.get_int( 0 );
hi = jarr.get_int( 1 );
if( lo >= hi ) {
jarr.throw_error( "uniform array should be in order [ lo, hi ]" );
}
if( lo < 0 || hi < 0 ) {
jarr.throw_error( "both elements must be 0 or greater" );
}
impl_ = make_shared_fast<uniform_distribution>( lo, hi );
} else if( jin.test_object() ) {
JsonObject jo = jin.get_object();
if( jo.has_member( "bounds" ) ) {
if( jo.get_array( "bounds" ).size() != 2 ) {
jo.throw_error( "bounds array has wrong number of elements, should be [ lo, hi ]" );
}
lo = jo.get_array( "bounds" ).get_int( 0 );
hi = jo.get_array( "bounds" ).get_int( 1 );
if( hi < 0 ) {
hi = INT_MAX;
}
if( lo < 0 ) {
lo = 0;
}
if( lo >= hi ) {
jo.throw_error( "bounds array should be in order [ lo, hi ]" );
}
}
if( jo.has_member( "poisson" ) ) {
if( jo.has_member( "binomial" ) ) {
jo.throw_error( "can't use multiple distribution types" );
}
double mean = jo.get_float( "poisson", 1.0 );
if( mean <= 0.0 ) {
jo.throw_error( "poisson mean must be greater than 0.0" );
}
impl_ = make_shared_fast<poisson_distribution>( lo, hi, mean );
} else if( jo.has_member( "binomial" ) ) {
JsonArray arr = jo.get_array( "binomial" );
if( jo.get_array( "binomial" ).size() != 2 ) {
jo.throw_error( "binomial array has wrong number of elements, should be [ t, p ]" );
}
int t = jo.get_array( "binomial" ).get_int( 0 );
double p = jo.get_array( "binomial" ).get_float( 1 );
if( t < 0 ) {
jo.throw_error( "t trials must be 0 or greater" );
}
if( p < 0.0 || p > 1.0 ) {
jo.throw_error( "success probability must be between 0.0 and 1.0" );
}
impl_ = make_shared_fast<binomial_distribution>( lo, hi, t, p );
} else {
jo.throw_error( R"(Expected "poisson" or "binomial" member)" );
}
} else {
jin.throw_error( "expected integer, array or object" );
}
}
|
