path: root/devdocs/go/math%2Frand%2Findex.html

<h1> Package rand  </h1>     <ul id="short-nav">
<li><code>import "math/rand"</code></li>
<li><a href="#pkg-overview" class="overviewLink">Overview</a></li>
<li><a href="#pkg-index" class="indexLink">Index</a></li>
<li><a href="#pkg-examples" class="examplesLink">Examples</a></li>
<li><a href="#pkg-subdirectories">Subdirectories</a></li>
</ul>     <h2 id="pkg-overview">Overview </h2> <p>Package rand implements pseudo-random number generators suitable for tasks such as simulation, but it should not be used for security-sensitive work. </p>
<p>Random numbers are generated by a <a href="#Source">Source</a>, usually wrapped in a <a href="#Rand">Rand</a>. Both types should be used by a single goroutine at a time: sharing among multiple goroutines requires some kind of synchronization. </p>
<p>Top-level functions, such as <a href="#Float64">Float64</a> and <a href="#Int">Int</a>, are safe for concurrent use by multiple goroutines. </p>
<p>This package's outputs might be easily predictable regardless of how it's seeded. For random numbers suitable for security-sensitive work, see the crypto/rand package. </p>   <h4 id="example_"> <span class="text">Example</span>
</h4> <p>Code:</p> <pre class="code" data-language="go">
answers := []string{
    "It is certain",
    "It is decidedly so",
    "Without a doubt",
    "Yes definitely",
    "You may rely on it",
    "As I see it yes",
    "Most likely",
    "Outlook good",
    "Yes",
    "Signs point to yes",
    "Reply hazy try again",
    "Ask again later",
    "Better not tell you now",
    "Cannot predict now",
    "Concentrate and ask again",
    "Don't count on it",
    "My reply is no",
    "My sources say no",
    "Outlook not so good",
    "Very doubtful",
}
fmt.Println("Magic 8-Ball says:", answers[rand.Intn(len(answers))])
</pre>      <h4 id="example__rand"> <span class="text">Example (Rand)</span>
</h4> <p>This example shows the use of each of the methods on a *Rand. The use of the global functions is the same, without the receiver. </p> <p>Code:</p> <pre class="code" data-language="go">// Create and seed the generator.
// Typically a non-fixed seed should be used, such as time.Now().UnixNano().
// Using a fixed seed will produce the same output on every run.
r := rand.New(rand.NewSource(99))

// The tabwriter here helps us generate aligned output.
w := tabwriter.NewWriter(os.Stdout, 1, 1, 1, ' ', 0)
defer w.Flush()
show := func(name string, v1, v2, v3 any) {
    fmt.Fprintf(w, "%s\t%v\t%v\t%v\n", name, v1, v2, v3)
}

// Float32 and Float64 values are in [0, 1).
show("Float32", r.Float32(), r.Float32(), r.Float32())
show("Float64", r.Float64(), r.Float64(), r.Float64())

// ExpFloat64 values have an average of 1 but decay exponentially.
show("ExpFloat64", r.ExpFloat64(), r.ExpFloat64(), r.ExpFloat64())

// NormFloat64 values have an average of 0 and a standard deviation of 1.
show("NormFloat64", r.NormFloat64(), r.NormFloat64(), r.NormFloat64())

// Int31, Int63, and Uint32 generate values of the given width.
// The Int method (not shown) is like either Int31 or Int63
// depending on the size of 'int'.
show("Int31", r.Int31(), r.Int31(), r.Int31())
show("Int63", r.Int63(), r.Int63(), r.Int63())
show("Uint32", r.Uint32(), r.Uint32(), r.Uint32())

// Intn, Int31n, and Int63n limit their output to be &lt; n.
// They do so more carefully than using r.Int()%n.
show("Intn(10)", r.Intn(10), r.Intn(10), r.Intn(10))
show("Int31n(10)", r.Int31n(10), r.Int31n(10), r.Int31n(10))
show("Int63n(10)", r.Int63n(10), r.Int63n(10), r.Int63n(10))

// Perm generates a random permutation of the numbers [0, n).
show("Perm", r.Perm(5), r.Perm(5), r.Perm(5))
</pre> <p>Output:</p> <pre class="output" data-language="go">Float32     0.2635776           0.6358173           0.6718283
Float64     0.628605430454327   0.4504798828572669  0.9562755949377957
ExpFloat64  0.3362240648200941  1.4256072328483647  0.24354758816173044
NormFloat64 0.17233959114940064 1.577014951434847   0.04259129641113857
Int31       1501292890          1486668269          182840835
Int63       3546343826724305832 5724354148158589552 5239846799706671610
Uint32      2760229429          296659907           1922395059
Intn(10)    1                   2                   5
Int31n(10)  4                   7                   8
Int63n(10)  7                   6                   3
Perm        [1 4 2 3 0]         [4 2 1 3 0]         [1 2 4 0 3]
</pre>        <h2 id="pkg-index">Index </h2>  <ul id="manual-nav">
<li><a href="#ExpFloat64">func ExpFloat64() float64</a></li>
<li><a href="#Float32">func Float32() float32</a></li>
<li><a href="#Float64">func Float64() float64</a></li>
<li><a href="#Int">func Int() int</a></li>
<li><a href="#Int31">func Int31() int32</a></li>
<li><a href="#Int31n">func Int31n(n int32) int32</a></li>
<li><a href="#Int63">func Int63() int64</a></li>
<li><a href="#Int63n">func Int63n(n int64) int64</a></li>
<li><a href="#Intn">func Intn(n int) int</a></li>
<li><a href="#NormFloat64">func NormFloat64() float64</a></li>
<li><a href="#Perm">func Perm(n int) []int</a></li>
<li><a href="#Read">func Read(p []byte) (n int, err error)</a></li>
<li><a href="#Seed">func Seed(seed int64)</a></li>
<li><a href="#Shuffle">func Shuffle(n int, swap func(i, j int))</a></li>
<li><a href="#Uint32">func Uint32() uint32</a></li>
<li><a href="#Uint64">func Uint64() uint64</a></li>
<li><a href="#Rand">type Rand</a></li>
<li> <a href="#New">func New(src Source) *Rand</a>
</li>
<li> <a href="#Rand.ExpFloat64">func (r *Rand) ExpFloat64() float64</a>
</li>
<li> <a href="#Rand.Float32">func (r *Rand) Float32() float32</a>
</li>
<li> <a href="#Rand.Float64">func (r *Rand) Float64() float64</a>
</li>
<li> <a href="#Rand.Int">func (r *Rand) Int() int</a>
</li>
<li> <a href="#Rand.Int31">func (r *Rand) Int31() int32</a>
</li>
<li> <a href="#Rand.Int31n">func (r *Rand) Int31n(n int32) int32</a>
</li>
<li> <a href="#Rand.Int63">func (r *Rand) Int63() int64</a>
</li>
<li> <a href="#Rand.Int63n">func (r *Rand) Int63n(n int64) int64</a>
</li>
<li> <a href="#Rand.Intn">func (r *Rand) Intn(n int) int</a>
</li>
<li> <a href="#Rand.NormFloat64">func (r *Rand) NormFloat64() float64</a>
</li>
<li> <a href="#Rand.Perm">func (r *Rand) Perm(n int) []int</a>
</li>
<li> <a href="#Rand.Read">func (r *Rand) Read(p []byte) (n int, err error)</a>
</li>
<li> <a href="#Rand.Seed">func (r *Rand) Seed(seed int64)</a>
</li>
<li> <a href="#Rand.Shuffle">func (r *Rand) Shuffle(n int, swap func(i, j int))</a>
</li>
<li> <a href="#Rand.Uint32">func (r *Rand) Uint32() uint32</a>
</li>
<li> <a href="#Rand.Uint64">func (r *Rand) Uint64() uint64</a>
</li>
<li><a href="#Source">type Source</a></li>
<li> <a href="#NewSource">func NewSource(seed int64) Source</a>
</li>
<li><a href="#Source64">type Source64</a></li>
<li><a href="#Zipf">type Zipf</a></li>
<li> <a href="#NewZipf">func NewZipf(r *Rand, s float64, v float64, imax uint64) *Zipf</a>
</li>
<li> <a href="#Zipf.Uint64">func (z *Zipf) Uint64() uint64</a>
</li>
</ul> <div id="pkg-examples"> <h3>Examples</h3>  <dl> <dd><a class="exampleLink" href="#example_">Package</a></dd> <dd><a class="exampleLink" href="#example_Intn">Intn</a></dd> <dd><a class="exampleLink" href="#example_Perm">Perm</a></dd> <dd><a class="exampleLink" href="#example_Shuffle">Shuffle</a></dd> <dd><a class="exampleLink" href="#example_Shuffle_slicesInUnison">Shuffle (SlicesInUnison)</a></dd> <dd><a class="exampleLink" href="#example__rand">Package (Rand)</a></dd> </dl> </div> <h3>Package files</h3> <p>  <span>exp.go</span> <span>normal.go</span> <span>rand.go</span> <span>rng.go</span> <span>zipf.go</span>  </p>   <h2 id="ExpFloat64">func <span>ExpFloat64</span>  </h2> <pre data-language="go">func ExpFloat64() float64</pre> <p>ExpFloat64 returns an exponentially distributed float64 in the range (0, +[math.MaxFloat64]] with an exponential distribution whose rate parameter (lambda) is 1 and whose mean is 1/lambda (1) from the default <a href="#Source">Source</a>. To produce a distribution with a different rate parameter, callers can adjust the output using: </p>
<pre data-language="go">sample = ExpFloat64() / desiredRateParameter
</pre> <h2 id="Float32">func <span>Float32</span>  </h2> <pre data-language="go">func Float32() float32</pre> <p>Float32 returns, as a float32, a pseudo-random number in the half-open interval [0.0,1.0) from the default <a href="#Source">Source</a>. </p>
<h2 id="Float64">func <span>Float64</span>  </h2> <pre data-language="go">func Float64() float64</pre> <p>Float64 returns, as a float64, a pseudo-random number in the half-open interval [0.0,1.0) from the default <a href="#Source">Source</a>. </p>
<h2 id="Int">func <span>Int</span>  </h2> <pre data-language="go">func Int() int</pre> <p>Int returns a non-negative pseudo-random int from the default <a href="#Source">Source</a>. </p>
<h2 id="Int31">func <span>Int31</span>  </h2> <pre data-language="go">func Int31() int32</pre> <p>Int31 returns a non-negative pseudo-random 31-bit integer as an int32 from the default <a href="#Source">Source</a>. </p>
<h2 id="Int31n">func <span>Int31n</span>  </h2> <pre data-language="go">func Int31n(n int32) int32</pre> <p>Int31n returns, as an int32, a non-negative pseudo-random number in the half-open interval [0,n) from the default <a href="#Source">Source</a>. It panics if n &lt;= 0. </p>
<h2 id="Int63">func <span>Int63</span>  </h2> <pre data-language="go">func Int63() int64</pre> <p>Int63 returns a non-negative pseudo-random 63-bit integer as an int64 from the default <a href="#Source">Source</a>. </p>
<h2 id="Int63n">func <span>Int63n</span>  </h2> <pre data-language="go">func Int63n(n int64) int64</pre> <p>Int63n returns, as an int64, a non-negative pseudo-random number in the half-open interval [0,n) from the default <a href="#Source">Source</a>. It panics if n &lt;= 0. </p>
<h2 id="Intn">func <span>Intn</span>  </h2> <pre data-language="go">func Intn(n int) int</pre> <p>Intn returns, as an int, a non-negative pseudo-random number in the half-open interval [0,n) from the default <a href="#Source">Source</a>. It panics if n &lt;= 0. </p>   <h4 id="example_Intn"> <span class="text">Example</span>
</h4> <p>Code:</p> <pre class="code" data-language="go">
fmt.Println(rand.Intn(100))
fmt.Println(rand.Intn(100))
fmt.Println(rand.Intn(100))
</pre>   <h2 id="NormFloat64">func <span>NormFloat64</span>  </h2> <pre data-language="go">func NormFloat64() float64</pre> <p>NormFloat64 returns a normally distributed float64 in the range [-<span>math.MaxFloat64</span>, +[math.MaxFloat64]] with standard normal distribution (mean = 0, stddev = 1) from the default <a href="#Source">Source</a>. To produce a different normal distribution, callers can adjust the output using: </p>
<pre data-language="go">sample = NormFloat64() * desiredStdDev + desiredMean
</pre> <h2 id="Perm">func <span>Perm</span>  </h2> <pre data-language="go">func Perm(n int) []int</pre> <p>Perm returns, as a slice of n ints, a pseudo-random permutation of the integers in the half-open interval [0,n) from the default <a href="#Source">Source</a>. </p>   <h4 id="example_Perm"> <span class="text">Example</span>
</h4> <p>Code:</p> <pre class="code" data-language="go">for _, value := range rand.Perm(3) {
    fmt.Println(value)
}

</pre> <p>Output:</p> <pre class="output" data-language="go">1
2
0
</pre>   <h2 id="Read">func <span>Read</span>  <span title="Added in Go 1.6">1.6</span> </h2> <pre data-language="go">func Read(p []byte) (n int, err error)</pre> <p>Read generates len(p) random bytes from the default <a href="#Source">Source</a> and writes them into p. It always returns len(p) and a nil error. Read, unlike the <a href="#Rand.Read">Rand.Read</a> method, is safe for concurrent use. </p>
<p>Deprecated: For almost all use cases, <span>crypto/rand.Read</span> is more appropriate. </p>
<h2 id="Seed">func <span>Seed</span>  </h2> <pre data-language="go">func Seed(seed int64)</pre> <p>Seed uses the provided seed value to initialize the default Source to a deterministic state. Seed values that have the same remainder when divided by 2³¹-1 generate the same pseudo-random sequence. Seed, unlike the <a href="#Rand.Seed">Rand.Seed</a> method, is safe for concurrent use. </p>
<p>If Seed is not called, the generator is seeded randomly at program startup. </p>
<p>Prior to Go 1.20, the generator was seeded like Seed(1) at program startup. To force the old behavior, call Seed(1) at program startup. Alternately, set GODEBUG=randautoseed=0 in the environment before making any calls to functions in this package. </p>
<p>Deprecated: As of Go 1.20 there is no reason to call Seed with a random value. Programs that call Seed with a known value to get a specific sequence of results should use New(NewSource(seed)) to obtain a local random generator. </p>
<h2 id="Shuffle">func <span>Shuffle</span>  <span title="Added in Go 1.10">1.10</span> </h2> <pre data-language="go">func Shuffle(n int, swap func(i, j int))</pre> <p>Shuffle pseudo-randomizes the order of elements using the default <a href="#Source">Source</a>. n is the number of elements. Shuffle panics if n &lt; 0. swap swaps the elements with indexes i and j. </p>   <h4 id="example_Shuffle"> <span class="text">Example</span>
</h4> <p>Code:</p> <pre class="code" data-language="go">
words := strings.Fields("ink runs from the corners of my mouth")
rand.Shuffle(len(words), func(i, j int) {
    words[i], words[j] = words[j], words[i]
})
fmt.Println(words)
</pre>      <h4 id="example_Shuffle_slicesInUnison"> <span class="text">Example (SlicesInUnison)</span>
</h4> <p>Code:</p> <pre class="code" data-language="go">
numbers := []byte("12345")
letters := []byte("ABCDE")
// Shuffle numbers, swapping corresponding entries in letters at the same time.
rand.Shuffle(len(numbers), func(i, j int) {
    numbers[i], numbers[j] = numbers[j], numbers[i]
    letters[i], letters[j] = letters[j], letters[i]
})
for i := range numbers {
    fmt.Printf("%c: %c\n", letters[i], numbers[i])
}
</pre>   <h2 id="Uint32">func <span>Uint32</span>  </h2> <pre data-language="go">func Uint32() uint32</pre> <p>Uint32 returns a pseudo-random 32-bit value as a uint32 from the default <a href="#Source">Source</a>. </p>
<h2 id="Uint64">func <span>Uint64</span>  <span title="Added in Go 1.8">1.8</span> </h2> <pre data-language="go">func Uint64() uint64</pre> <p>Uint64 returns a pseudo-random 64-bit value as a uint64 from the default <a href="#Source">Source</a>. </p>
<h2 id="Rand">type <span>Rand</span>  </h2> <p>A Rand is a source of random numbers. </p>
<pre data-language="go">type Rand struct {
    // contains filtered or unexported fields
}
</pre> <h3 id="New">func <span>New</span>  </h3> <pre data-language="go">func New(src Source) *Rand</pre> <p>New returns a new <a href="#Rand">Rand</a> that uses random values from src to generate other random values. </p>
<h3 id="Rand.ExpFloat64">func (*Rand) <span>ExpFloat64</span>  </h3> <pre data-language="go">func (r *Rand) ExpFloat64() float64</pre> <p>ExpFloat64 returns an exponentially distributed float64 in the range (0, +[math.MaxFloat64]] with an exponential distribution whose rate parameter (lambda) is 1 and whose mean is 1/lambda (1). To produce a distribution with a different rate parameter, callers can adjust the output using: </p>
<pre data-language="go">sample = ExpFloat64() / desiredRateParameter
</pre> <h3 id="Rand.Float32">func (*Rand) <span>Float32</span>  </h3> <pre data-language="go">func (r *Rand) Float32() float32</pre> <p>Float32 returns, as a float32, a pseudo-random number in the half-open interval [0.0,1.0). </p>
<h3 id="Rand.Float64">func (*Rand) <span>Float64</span>  </h3> <pre data-language="go">func (r *Rand) Float64() float64</pre> <p>Float64 returns, as a float64, a pseudo-random number in the half-open interval [0.0,1.0). </p>
<h3 id="Rand.Int">func (*Rand) <span>Int</span>  </h3> <pre data-language="go">func (r *Rand) Int() int</pre> <p>Int returns a non-negative pseudo-random int. </p>
<h3 id="Rand.Int31">func (*Rand) <span>Int31</span>  </h3> <pre data-language="go">func (r *Rand) Int31() int32</pre> <p>Int31 returns a non-negative pseudo-random 31-bit integer as an int32. </p>
<h3 id="Rand.Int31n">func (*Rand) <span>Int31n</span>  </h3> <pre data-language="go">func (r *Rand) Int31n(n int32) int32</pre> <p>Int31n returns, as an int32, a non-negative pseudo-random number in the half-open interval [0,n). It panics if n &lt;= 0. </p>
<h3 id="Rand.Int63">func (*Rand) <span>Int63</span>  </h3> <pre data-language="go">func (r *Rand) Int63() int64</pre> <p>Int63 returns a non-negative pseudo-random 63-bit integer as an int64. </p>
<h3 id="Rand.Int63n">func (*Rand) <span>Int63n</span>  </h3> <pre data-language="go">func (r *Rand) Int63n(n int64) int64</pre> <p>Int63n returns, as an int64, a non-negative pseudo-random number in the half-open interval [0,n). It panics if n &lt;= 0. </p>
<h3 id="Rand.Intn">func (*Rand) <span>Intn</span>  </h3> <pre data-language="go">func (r *Rand) Intn(n int) int</pre> <p>Intn returns, as an int, a non-negative pseudo-random number in the half-open interval [0,n). It panics if n &lt;= 0. </p>
<h3 id="Rand.NormFloat64">func (*Rand) <span>NormFloat64</span>  </h3> <pre data-language="go">func (r *Rand) NormFloat64() float64</pre> <p>NormFloat64 returns a normally distributed float64 in the range -<span>math.MaxFloat64</span> through +[math.MaxFloat64] inclusive, with standard normal distribution (mean = 0, stddev = 1). To produce a different normal distribution, callers can adjust the output using: </p>
<pre data-language="go">sample = NormFloat64() * desiredStdDev + desiredMean
</pre> <h3 id="Rand.Perm">func (*Rand) <span>Perm</span>  </h3> <pre data-language="go">func (r *Rand) Perm(n int) []int</pre> <p>Perm returns, as a slice of n ints, a pseudo-random permutation of the integers in the half-open interval [0,n). </p>
<h3 id="Rand.Read">func (*Rand) <span>Read</span>  <span title="Added in Go 1.6">1.6</span> </h3> <pre data-language="go">func (r *Rand) Read(p []byte) (n int, err error)</pre> <p>Read generates len(p) random bytes and writes them into p. It always returns len(p) and a nil error. Read should not be called concurrently with any other Rand method. </p>
<h3 id="Rand.Seed">func (*Rand) <span>Seed</span>  </h3> <pre data-language="go">func (r *Rand) Seed(seed int64)</pre> <p>Seed uses the provided seed value to initialize the generator to a deterministic state. Seed should not be called concurrently with any other <a href="#Rand">Rand</a> method. </p>
<h3 id="Rand.Shuffle">func (*Rand) <span>Shuffle</span>  <span title="Added in Go 1.10">1.10</span> </h3> <pre data-language="go">func (r *Rand) Shuffle(n int, swap func(i, j int))</pre> <p>Shuffle pseudo-randomizes the order of elements. n is the number of elements. Shuffle panics if n &lt; 0. swap swaps the elements with indexes i and j. </p>
<h3 id="Rand.Uint32">func (*Rand) <span>Uint32</span>  </h3> <pre data-language="go">func (r *Rand) Uint32() uint32</pre> <p>Uint32 returns a pseudo-random 32-bit value as a uint32. </p>
<h3 id="Rand.Uint64">func (*Rand) <span>Uint64</span>  <span title="Added in Go 1.8">1.8</span> </h3> <pre data-language="go">func (r *Rand) Uint64() uint64</pre> <p>Uint64 returns a pseudo-random 64-bit value as a uint64. </p>
<h2 id="Source">type <span>Source</span>  </h2> <p>A Source represents a source of uniformly-distributed pseudo-random int64 values in the range [0, 1&lt;&lt;63). </p>
<p>A Source is not safe for concurrent use by multiple goroutines. </p>
<pre data-language="go">type Source interface {
    Int63() int64
    Seed(seed int64)
}</pre> <h3 id="NewSource">func <span>NewSource</span>  </h3> <pre data-language="go">func NewSource(seed int64) Source</pre> <p>NewSource returns a new pseudo-random <a href="#Source">Source</a> seeded with the given value. Unlike the default <a href="#Source">Source</a> used by top-level functions, this source is not safe for concurrent use by multiple goroutines. The returned <a href="#Source">Source</a> implements <a href="#Source64">Source64</a>. </p>
<h2 id="Source64">type <span>Source64</span>  <span title="Added in Go 1.8">1.8</span> </h2> <p>A Source64 is a <a href="#Source">Source</a> that can also generate uniformly-distributed pseudo-random uint64 values in the range [0, 1&lt;&lt;64) directly. If a <a href="#Rand">Rand</a> r's underlying <a href="#Source">Source</a> s implements Source64, then r.Uint64 returns the result of one call to s.Uint64 instead of making two calls to s.Int63. </p>
<pre data-language="go">type Source64 interface {
    Source
    Uint64() uint64
}</pre> <h2 id="Zipf">type <span>Zipf</span>  </h2> <p>A Zipf generates Zipf distributed variates. </p>
<pre data-language="go">type Zipf struct {
    // contains filtered or unexported fields
}
</pre> <h3 id="NewZipf">func <span>NewZipf</span>  </h3> <pre data-language="go">func NewZipf(r *Rand, s float64, v float64, imax uint64) *Zipf</pre> <p>NewZipf returns a <a href="#Zipf">Zipf</a> variate generator. The generator generates values k ∈ [0, imax] such that P(k) is proportional to (v + k) ** (-s). Requirements: s &gt; 1 and v &gt;= 1. </p>
<h3 id="Zipf.Uint64">func (*Zipf) <span>Uint64</span>  </h3> <pre data-language="go">func (z *Zipf) Uint64() uint64</pre> <p>Uint64 returns a value drawn from the <a href="#Zipf">Zipf</a> distribution described by the <a href="#Zipf">Zipf</a> object. </p>
<h2 id="pkg-subdirectories">Subdirectories</h2> <div class="pkg-dir"> <table> <tr> <th class="pkg-name">Name</th> <th class="pkg-synopsis">Synopsis</th> </tr> <tr> <td colspan="2"><a href="../index">..</a></td> </tr> <tr> <td class="pkg-name"> <a href="v2/index">v2</a> </td> <td class="pkg-synopsis"> Package rand implements pseudo-random number generators suitable for tasks such as simulation, but it should not be used for security-sensitive work. </td> </tr> </table> </div><div class="_attribution">
  <p class="_attribution-p">
    &copy; Google, Inc.<br>Licensed under the Creative Commons Attribution License 3.0.<br>
    <a href="http://golang.org/pkg/math/rand/" class="_attribution-link">http://golang.org/pkg/math/rand/</a>
  </p>
</div>