/// <summary>Preprocess data set to create dictionary.</summary>
/// <param name="examples">.</param>
public override void PreProcess(IEnumerable <object> examples)
{
var q = from s in examples select Ject.Get(s, this.Name).ToString();
if (this.AsEnum)
{
this.Dictionary = StringHelpers.BuildEnumDictionary(q).Select(kv => kv.Key).ToArray();
}
else
{
switch (this.SplitType)
{
case StringSplitType.Character:
this.Dictionary =
StringHelpers.BuildCharDictionary(q, this.Exclude).Select(kv => kv.Key).ToArray();
break;
case StringSplitType.Word:
this.Dictionary =
StringHelpers.BuildWordDictionary(q, this.Separator, this.Exclude)
.Select(kv => kv.Key)
.ToArray();
break;
}
}
}
public void Test_Vector_Conversion_Simple_Numbers_And_Chars_As_Enum()
{
Descriptor d = new Descriptor();
var dictionary = StringHelpers.BuildCharDictionary(ShortStrings)
.Select(k => k.Key)
.ToArray();
d.Features = new Property[]
{
new Property {
Name = "Age"
},
new Property {
Name = "Height"
},
new StringProperty {
Name = "Chars",
Dictionary = dictionary,
SplitType = StringSplitType.Character,
AsEnum = true
},
new Property {
Name = "Weight"
},
new Property {
Name = "Good"
},
};
// ["O", "N", "E", "#SYM#", "T", "W", "#NUM#", "H", "R"]
// [ 2, 1, 0, 1, 0, 0, 1, 0, 3]
var o = new { Age = 23, Height = 6.21d, Chars = "N", Weight = 220m, Good = false, };
// array generated by descriptor ordering
var truths = new double[] { 23, 6.21,
/* BEGIN CHARS */
1,
/* END CHARS */
220, -1 };
var actual = d.Convert(o);
Assert.AreEqual(truths, actual);
o = new { Age = 23, Height = 6.21d, Chars = "!", Weight = 220m, Good = false, };
// array generated by descriptor ordering
truths = new double[] { 23, 6.21,
/* BEGIN CHARS */
3,
/* END CHARS */
220, -1 };
actual = d.Convert(o);
Assert.AreEqual(truths, actual);
}
public void Test_Char_Dictionary_Gen()
{
var d = StringHelpers.BuildCharDictionary(ShortStrings);
// should have the following:
// O(2), N(1), E(5), #SYM#(5), T(3), W(1), #NUM#(1), H(2), R(2)
var dict = new Dictionary <string, double>
{
{ "O", 2 },
{ "N", 1 },
{ "E", 5 },
{ "#SYM#", 5 },
{ "T", 3 },
{ "W", 1 },
{ "#NUM#", 1 },
{ "H", 2 },
{ "R", 2 },
};
Assert.AreEqual(dict, d);
}