A distribution over random integers.
Declared in <absl/random/discrete_distribution.h>
template<typename IntType = int>
class discrete_distribution;A discrete distribution produces random integers i, where 0 <= i < n distributed according to the discrete probability function:
P(i|p0,...,pn-1)=pi
This class is an implementation of discrete_distribution (see [rand.dist.samp.discrete]).
The algorithm used is Walker's Aliasing algorithm, described in Knuth, Vol 2. absl::discrete_distribution takes O(N) time to precompute the probabilities (where N is the number of possible outcomes in the distribution) at construction, and then takes O(1) time for each variate generation. Many other implementations also take O(N) time to construct an ordered sequence of partial sums, plus O(log N) time per variate to binary search.
| Name | Description |
|---|---|
param_type | The parameter set of the distribution. |
| Name | Description |
|---|---|
result_type | The type of the values produced by the distribution. |
| Name | Description |
|---|---|
discrete_distribution [constructor] | Constructors |
max | Returns the largest value the distribution can produce. |
min | Returns the smallest value the distribution can produce. |
operator() | Function call operators |
param | param overloads |
probabilities | Returns the normalized probabilities of the outcomes. |
reset | Resets the internal state of the distribution. |
| Name | Description |
|---|---|
absl::operator!= | Compares two distributions for inequality. |
absl::operator== | Compares two distributions for equality. |