public static StringDistribution SubAverageConditional(StringDistribution str, int start, int minLength, int maxLength)
{
Argument.CheckIfNotNull(str, "str");
Argument.CheckIfInRange(start >= 0, "start", "Start index must be non-negative.");
Argument.CheckIfInRange(minLength >= 0, "minLength", "Min length must be non-negative.");
Argument.CheckIfInRange(maxLength >= 0, "maxLength", "Max length must be non-negative.");
if (str.IsPointMass)
{
var strPoint = str.Point;
var alts = new HashSet <string>();
for (int length = minLength; length <= maxLength; length++)
{
var s = strPoint.Substring(start, Math.Min(length, strPoint.Length));
alts.Add(s);
}
return(StringDistribution.OneOf(alts));
}
var anyChar = StringAutomaton.ConstantOnElement(1.0, DiscreteChar.Any());
var transducer = StringTransducer.Consume(StringAutomaton.Repeat(anyChar, minTimes: start, maxTimes: start));
transducer.AppendInPlace(StringTransducer.Copy(StringAutomaton.Repeat(anyChar, minTimes: minLength, maxTimes: maxLength)));
transducer.AppendInPlace(StringTransducer.Consume(StringAutomaton.Constant(1.0)));
return(StringDistribution.FromWorkspace(transducer.ProjectSource(str.GetWorkspaceOrPoint())));
}
/// <summary>EP message to <c>str</c>.</summary>
/// <param name="sub">Incoming message from <c>sub</c>.</param>
/// <param name="start">Constant value for <c>start</c>.</param>
/// <param name="length">Constant value for <c>length</c>.</param>
/// <returns>The outgoing EP message to the <c>str</c> argument.</returns>
/// <remarks>
/// <para>The outgoing message is a distribution matching the moments of <c>str</c> as the random arguments are varied. The formula is <c>proj[p(str) sum_(sub) p(sub) factor(sub,str,start,length)]/p(str)</c>.</para>
/// </remarks>
public static StringDistribution StrAverageConditional(StringDistribution sub, int start, int length)
{
Argument.CheckIfNotNull(sub, "sub");
Argument.CheckIfInRange(start >= 0, "start", "Start index must be non-negative.");
Argument.CheckIfInRange(length >= 0, "length", "Length must be non-negative.");
var anyChar = StringAutomaton.ConstantOnElement(1.0, DiscreteChar.Any());
var transducer = StringTransducer.Produce(StringAutomaton.Repeat(anyChar, minTimes: start, maxTimes: start));
transducer.AppendInPlace(StringTransducer.Copy(StringAutomaton.Repeat(anyChar, minTimes: length, maxTimes: length)));
transducer.AppendInPlace(StringTransducer.Produce(StringAutomaton.Constant(1.0)));
return(StringDistribution.FromWorkspace(transducer.ProjectSource(sub)));
}
public void StringFromItems()
{
Variable <string> str = Variable.StringOfLength(2);
Variable <char> char1 = Variable.GetItem(str, 0);
Variable <char> char2 = Variable.GetItem(str, 1);
var engine = new InferenceEngine();
char1.ObservedValue = 'a';
char2.ObservedValue = 'b';
Assert.Equal(StringDistribution.PointMass("ab"), engine.Infer <StringDistribution>(str), ProbEps);
char1.ClearObservedValue();
Assert.Equal(
StringDistribution.Char(DiscreteChar.Any()) + StringDistribution.Char('b'),
engine.Infer <StringDistribution>(str),
ProbEps);
}
/// <summary> /// Creates a uniform distribution over any string starting and ending with a non-word character (possibly of length 1). /// </summary> /// <returns>The created distribution.</returns> public static StringDistribution WordMiddle() => WordMiddle(DiscreteChar.Any());
/// <summary> /// Creates a uniform distribution over either the empty string or any string starting with a non-word character. /// </summary> /// <returns>The created distribution.</returns> public static StringDistribution WordSuffix() => WordSuffix(DiscreteChar.Any());
/// <summary> /// Creates a uniform distribution over any string starting and ending with a non-word character, /// with a length in given bounds. /// </summary> /// <param name="minLength">The minimum allowed string length.</param> /// <param name="maxLength">The maximum allowed string length.</param> /// <returns>The created distribution.</returns> public static StringDistribution WordMiddle(int minLength, int maxLength) => WordMiddle(minLength, maxLength, DiscreteChar.Any());