public virtual void IncrementCount(K1 o1, K2 o2, double count)
{
ClassicCounter <K2> c = GetCounter(o1);
c.IncrementCount(o2, count);
total += count;
}
public virtual void TestToSortedString()
{
ICounter <string> c = new ClassicCounter <string>();
c.SetCount("b", 0.25);
c.SetCount("a", 0.5);
c.SetCount("c", 1.0);
// check full argument version
string result = Counters.ToSortedString(c, 5, "%s%.1f", ":", "{%s}");
NUnit.Framework.Assert.AreEqual("{c1.0:a0.5:b0.3}", result);
// check version with no wrapper
result = Counters.ToSortedString(c, 2, "%2$f %1$s", "\n");
NUnit.Framework.Assert.AreEqual("1.000000 c\n0.500000 a", result);
// check some equivalences to other Counters methods
int k = 2;
result = Counters.ToSortedString(c, k, "%s=%s", ", ", "[%s]");
NUnit.Framework.Assert.AreEqual(Counters.ToString(c, k), result);
NUnit.Framework.Assert.AreEqual(Counters.ToBiggestValuesFirstString(c, k), result);
result = Counters.ToSortedString(c, k, "%2$g\t%1$s", "\n", "%s\n");
NUnit.Framework.Assert.AreEqual(Counters.ToVerticalString(c, k), result);
// test sorting by keys
result = Counters.ToSortedByKeysString(c, "%s=>%.2f", "; ", "<%s>");
NUnit.Framework.Assert.AreEqual("<a=>0.50; b=>0.25; c=>1.00>", result);
}
private double Percentage(OUT key, ClassicCounter <OUT> guessed, ClassicCounter <OUT> guessedCorrect)
{
double thisGuessed = guessed.GetCount(key);
double thisGuessedCorrect = guessedCorrect.GetCount(key);
return((thisGuessed == 0.0) ? 0.0 : thisGuessedCorrect / thisGuessed);
}
public virtual void TestSerializeStringCounter()
{
ICounter <string> counts = new ClassicCounter <string>();
for (int @base = -10; @base < 10; ++@base)
{
if (@base == 0)
{
continue;
}
for (int exponent = -100; exponent < 100; ++exponent)
{
double number = Math.Pow(Math.Pi * @base, exponent);
counts.SetCount(double.ToString(number), number);
}
}
File tmp = File.CreateTempFile("counts", ".tab.gz");
tmp.DeleteOnExit();
Counters.SerializeStringCounter(counts, tmp.GetPath());
ICounter <string> reread = Counters.DeserializeStringCounter(tmp.GetPath());
foreach (KeyValuePair <string, double> entry in reread.EntrySet())
{
double old = counts.GetCount(entry.Key);
NUnit.Framework.Assert.AreEqual(old, entry.Value, Math.Abs(old) / 1e5);
}
}
public virtual void TestGetDistributionFromLogValues()
{
ICounter <string> c1 = new ClassicCounter <string>();
c1.SetCount("p", 1.0);
c1.SetCount("q", 2.0);
c1.SetCount("r", 3.0);
c1.SetCount("s", 4.0);
// take log
Counters.LogInPlace(c1);
// now call distribution
Distribution <string> distribution = Distribution.GetDistributionFromLogValues(c1);
// test
NUnit.Framework.Assert.AreEqual(distribution.KeySet().Count, 4);
// size
// keys
NUnit.Framework.Assert.AreEqual(distribution.ContainsKey("p"), true);
NUnit.Framework.Assert.AreEqual(distribution.ContainsKey("q"), true);
NUnit.Framework.Assert.AreEqual(distribution.ContainsKey("r"), true);
NUnit.Framework.Assert.AreEqual(distribution.ContainsKey("s"), true);
// values
NUnit.Framework.Assert.AreEqual(distribution.GetCount("p"), 1.0E-1, 1E-10);
NUnit.Framework.Assert.AreEqual(distribution.GetCount("q"), 2.0E-1, 1E-10);
NUnit.Framework.Assert.AreEqual(distribution.GetCount("r"), 3.0E-1, 1E-10);
NUnit.Framework.Assert.AreEqual(distribution.GetCount("s"), 4.0E-1, 1E-10);
}
public DirichletProcess(IProbabilityDistribution <E> baseMeasure, double alpha)
{
this.baseMeasure = baseMeasure;
this.alpha = alpha;
this.sampled = new ClassicCounter <E>();
sampled.IncrementCount(null, alpha);
}
public virtual void SubtractAll(K1 key, ICounter <K2> c)
{
ClassicCounter <K2> myInner = GetCounter(key);
Counters.SubtractInPlace(myInner, c);
total -= c.TotalCount();
}
/// <summary>for testing purposes only</summary>
public static void Main(string[] args)
{
object[] a1 = new object[] { "a", "b" };
object[] a2 = new object[] { "a", "b" };
System.Console.Out.WriteLine(Arrays.Equals(a1, a2));
GeneralizedCounter <string> gc = new GeneralizedCounter <string>(3);
gc.IncrementCount(Arrays.AsList(new string[] { "a", "j", "x" }), 3.0);
gc.IncrementCount(Arrays.AsList(new string[] { "a", "l", "x" }), 3.0);
gc.IncrementCount(Arrays.AsList(new string[] { "b", "k", "y" }), 3.0);
gc.IncrementCount(Arrays.AsList(new string[] { "b", "k", "z" }), 3.0);
System.Console.Out.WriteLine("incremented counts.");
System.Console.Out.WriteLine(gc.DumpKeys());
System.Console.Out.WriteLine("string representation of generalized counter:");
System.Console.Out.WriteLine(gc.ToString());
gc.PrintKeySet();
System.Console.Out.WriteLine("entry set:\n" + gc.EntrySet());
ArrayPrintDouble(gc.GetCounts(Arrays.AsList(new string[] { "a", "j", "x" })));
ArrayPrintDouble(gc.GetCounts(Arrays.AsList(new string[] { "a", "j", "z" })));
ArrayPrintDouble(gc.GetCounts(Arrays.AsList(new string[] { "b", "k", "w" })));
ArrayPrintDouble(gc.GetCounts(Arrays.AsList(new string[] { "b", "k", "z" })));
GeneralizedCounter <string> gc1 = gc.Conditionalize(Arrays.AsList(new string[] { "a" }));
gc1.IncrementCount(Arrays.AsList(new string[] { "j", "x" }));
gc1.IncrementCount2D("j", "z");
GeneralizedCounter <string> gc2 = gc1.Conditionalize(Arrays.AsList(new string[] { "j" }));
gc2.IncrementCount1D("x");
System.Console.Out.WriteLine("Pretty-printing gc after incrementing gc1:");
gc.PrettyPrint();
System.Console.Out.WriteLine("Total: " + gc.TotalCount());
gc1.PrintKeySet();
System.Console.Out.WriteLine("another entry set:\n" + gc1.EntrySet());
ClassicCounter <IList <string> > c = gc.CounterView();
System.Console.Out.WriteLine("string representation of counter view:");
System.Console.Out.WriteLine(c.ToString());
double d1 = c.GetCount(Arrays.AsList(new string[] { "a", "j", "x" }));
double d2 = c.GetCount(Arrays.AsList(new string[] { "a", "j", "w" }));
System.Console.Out.WriteLine(d1 + " " + d2);
ClassicCounter <IList <string> > c1 = gc1.CounterView();
System.Console.Out.WriteLine("Count of {j,x} -- should be 3.0\t" + c1.GetCount(Arrays.AsList(new string[] { "j", "x" })));
System.Console.Out.WriteLine(c.KeySet() + " size " + c.KeySet().Count);
System.Console.Out.WriteLine(c1.KeySet() + " size " + c1.KeySet().Count);
System.Console.Out.WriteLine(c1.Equals(c));
System.Console.Out.WriteLine(c.Equals(c1));
System.Console.Out.WriteLine(c.Equals(c));
System.Console.Out.WriteLine("### testing equality of regular Counter...");
ClassicCounter <string> z1 = new ClassicCounter <string>();
ClassicCounter <string> z2 = new ClassicCounter <string>();
z1.IncrementCount("a1");
z1.IncrementCount("a2");
z2.IncrementCount("b");
System.Console.Out.WriteLine(z1.Equals(z2));
System.Console.Out.WriteLine(z1.ToString());
System.Console.Out.WriteLine(z1.KeySet().ToString());
}
public virtual Edu.Stanford.Nlp.Stats.Dirichlet <E> GetPosteriorDistribution(ICounter <E> counts)
{
ICounter <E> newParameters = new ClassicCounter <E>(parameters);
Counters.AddInPlace(newParameters, counts);
return(new Edu.Stanford.Nlp.Stats.Dirichlet <E>(newParameters));
}
public virtual void TestPointwiseMutualInformation()
{
ICounter <string> x = new ClassicCounter <string>();
x.IncrementCount("0", 0.8);
x.IncrementCount("1", 0.2);
ICounter <int> y = new ClassicCounter <int>();
y.IncrementCount(0, 0.25);
y.IncrementCount(1, 0.75);
ICounter <Pair <string, int> > joint;
joint = new ClassicCounter <Pair <string, int> >();
joint.IncrementCount(new Pair <string, int>("0", 0), 0.1);
joint.IncrementCount(new Pair <string, int>("0", 1), 0.7);
joint.IncrementCount(new Pair <string, int>("1", 0), 0.15);
joint.IncrementCount(new Pair <string, int>("1", 1), 0.05);
// Check that correct PMI values are calculated, using tables from
// http://en.wikipedia.org/wiki/Pointwise_mutual_information
double pmi;
Pair <string, int> pair;
pair = new Pair <string, int>("0", 0);
pmi = Counters.PointwiseMutualInformation(x, y, joint, pair);
NUnit.Framework.Assert.AreEqual(-1, pmi, 10e-5);
pair = new Pair <string, int>("0", 1);
pmi = Counters.PointwiseMutualInformation(x, y, joint, pair);
NUnit.Framework.Assert.AreEqual(0.222392421, pmi, 10e-5);
pair = new Pair <string, int>("1", 0);
pmi = Counters.PointwiseMutualInformation(x, y, joint, pair);
NUnit.Framework.Assert.AreEqual(1.584962501, pmi, 10e-5);
pair = new Pair <string, int>("1", 1);
pmi = Counters.PointwiseMutualInformation(x, y, joint, pair);
NUnit.Framework.Assert.AreEqual(-1.584962501, pmi, 10e-5);
}
public virtual void AddAll(K1 key, ICounter <K2> c)
{
ClassicCounter <K2> myInner = GetCounter(key);
Counters.AddInPlace(myInner, c);
total += c.TotalCount();
}
public virtual ClassicCounter <OUT> LastPrecision()
{
ClassicCounter <OUT> result = new ClassicCounter <OUT>();
Counters.AddInPlace(result, previousGuessedCorrect);
Counters.DivideInPlace(result, previousGuessed);
return(result);
}
public virtual ClassicCounter <OUT> LastRecall()
{
ClassicCounter <OUT> result = new ClassicCounter <OUT>();
Counters.AddInPlace(result, previousGoldCorrect);
Counters.DivideInPlace(result, previousGold);
return(result);
}
public virtual void SetCount(K1 o1, K2 o2, double count)
{
ClassicCounter <K2> c = GetCounter(o1);
double oldCount = GetCount(o1, o2);
total -= oldCount;
c.SetCount(o2, count);
total += count;
}
// it's empty, get rid of it!
public virtual void Remove(K1 key)
{
ClassicCounter <K2> counter = map[key];
if (counter != null)
{
total -= counter.TotalCount();
}
Sharpen.Collections.Remove(map, key);
}
/// <summary>
/// Returns the counters with keys as the first key and count as the
/// total count of the inner counter for that key
/// </summary>
/// <returns>counter of type K1</returns>
public virtual ICounter <K1> SumInnerCounter()
{
ICounter <K1> summed = new ClassicCounter <K1>();
foreach (K1 key in this.FirstKeySet())
{
summed.IncrementCount(key, this.GetCounter(key).TotalCount());
}
return(summed);
}
public virtual void AddAll(ITwoDimensionalCounterInterface <K1, K2> c)
{
foreach (K1 key in c.FirstKeySet())
{
ICounter <K2> inner = c.GetCounter(key);
ClassicCounter <K2> myInner = GetCounter(key);
Counters.AddInPlace(myInner, inner);
total += inner.TotalCount();
}
}
public virtual double GetCount(K1 o1, K2 o2)
{
ClassicCounter <K2> c = GetCounter(o1);
if (c.TotalCount() == 0.0 && !c.KeySet().Contains(o2))
{
return(DefaultReturnValue());
}
return(c.GetCount(o2));
}
public virtual bool ContainsKey(K1 o1, K2 o2)
{
if (!map.Contains(o1))
{
return(false);
}
ClassicCounter <K2> c = map[o1];
return(c.ContainsKey(o2));
}
public static double SampleBeta(double a, double b, Random random)
{
ICounter <bool> c = new ClassicCounter <bool>();
c.SetCount(true, a);
c.SetCount(false, b);
Multinomial <bool> beta = (new Edu.Stanford.Nlp.Stats.Dirichlet <bool>(c)).DrawSample(random);
return(beta.ProbabilityOf(true));
}
public virtual ClassicCounter <OUT> LastF1()
{
ClassicCounter <OUT> result = new ClassicCounter <OUT>();
ICollection <OUT> keys = Sets.Union(previousGuessed.KeySet(), previousGold.KeySet());
foreach (OUT key in keys)
{
result.SetCount(key, LastF1(key));
}
return(result);
}
/// <returns>total number of entries (key pairs)</returns>
public virtual int Size()
{
int result = 0;
foreach (K1 o in FirstKeySet())
{
ClassicCounter <K2> c = map[o];
result += c.Size();
}
return(result);
}
/// <returns>the inner Counter associated with key o</returns>
public virtual ClassicCounter <K2> GetCounter(K1 o)
{
ClassicCounter <K2> c = map[o];
if (c == null)
{
c = new ClassicCounter <K2>(innerMF);
c.SetDefaultReturnValue(defaultValue);
map[o] = c;
}
return(c);
}
public virtual void TestL2Norm()
{
ClassicCounter <string> c = new ClassicCounter <string>();
c.IncrementCount("a", 3);
c.IncrementCount("b", 4);
NUnit.Framework.Assert.AreEqual(5.0, Counters.L2Norm(c), Tolerance);
c.IncrementCount("c", 6);
c.IncrementCount("d", 4);
c.IncrementCount("e", 2);
NUnit.Framework.Assert.AreEqual(9.0, Counters.L2Norm(c), Tolerance);
}
/* Helper to simpleGoodTuringSmoothedCounter() */
private static ICounter <int> CollectCountCounts <E>(ICounter <E> counts)
{
ICounter <int> cc = new ClassicCounter <int>();
// counts of counts
foreach (KeyValuePair <E, double> entry in counts.EntrySet())
{
//E item = entry.getKey();
int count = (int)Math.Round(entry.Value);
cc.IncrementCount(count);
}
return(cc);
}
public virtual double Remove(K1 o1, K2 o2)
{
ClassicCounter <K2> c = GetCounter(o1);
double oldCount = GetCount(o1, o2);
total -= oldCount;
c.Remove(o2);
if (c.Size() == 0)
{
Sharpen.Collections.Remove(map, o1);
}
return(oldCount);
}
public virtual void SubtractAll(ITwoDimensionalCounterInterface <K1, K2> c, bool removeKeys)
{
foreach (K1 key in c.FirstKeySet())
{
ICounter <K2> inner = c.GetCounter(key);
ClassicCounter <K2> myInner = GetCounter(key);
Counters.SubtractInPlace(myInner, inner);
if (removeKeys)
{
Counters.RetainNonZeros(myInner);
}
total -= inner.TotalCount();
}
}
public virtual void RemoveZeroCounts()
{
ICollection <K1> firstKeySet = Generics.NewHashSet(FirstKeySet());
foreach (K1 k1 in firstKeySet)
{
ClassicCounter <K2> c = GetCounter(k1);
Counters.RetainNonZeros(c);
if (c.Size() == 0)
{
Sharpen.Collections.Remove(map, k1);
}
}
}
/// <summary>Returns a new Distribution<K> with counts averaged from the two given Distributions.</summary>
/// <remarks>
/// Returns a new Distribution<K> with counts averaged from the two given Distributions.
/// The average Distribution<K> will contain the union of keys in both
/// source Distributions, and each count will be the weighted average of the two source
/// counts for that key, a missing count in one Distribution
/// is treated as if it has probability equal to that returned by the probabilityOf() function.
/// </remarks>
/// <returns>
/// A new distribution with counts that are the mean of the resp. counts
/// in the given distributions with the remaining probability mass adjusted accordingly.
/// </returns>
public static Distribution <K> WeightedAverage <K>(Distribution <K> d1, double w1, Distribution <K> d2)
{
double w2 = 1.0 - w1;
ICollection <K> allKeys = GetSetOfAllKeys(d1, d2);
int numKeys = d1.GetNumberOfKeys();
ICounter <K> c = new ClassicCounter <K>();
foreach (K key in allKeys)
{
double newProbability = d1.ProbabilityOf(key) * w1 + d2.ProbabilityOf(key) * w2;
c.SetCount(key, newProbability);
}
return(Distribution.GetDistributionFromPartiallySpecifiedCounter(c, numKeys));
}
public virtual void TestL2Normalize()
{
ClassicCounter <string> c = new ClassicCounter <string>();
c.IncrementCount("a", 4.0);
c.IncrementCount("b", 2.0);
c.IncrementCount("c", 1.0);
c.IncrementCount("d", 2.0);
ICounter <string> d = Counters.L2Normalize(c);
NUnit.Framework.Assert.AreEqual(d.GetCount("a"), 0.8, Tolerance);
NUnit.Framework.Assert.AreEqual(d.GetCount("b"), 0.4, Tolerance);
NUnit.Framework.Assert.AreEqual(d.GetCount("c"), 0.2, Tolerance);
NUnit.Framework.Assert.AreEqual(d.GetCount("d"), 0.4, Tolerance);
}
