Changeset 17414 for branches/3040_VectorBasedGP/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification

Timestamp:

01/31/20 12:42:47 (5 years ago)

Author:

pfleck

Message:

#3040 Started adding UCI time series regression benchmarks.
Adapted parser (extracted format options & added parsing for double vectors).

File:

• branches/3040_VectorBasedGP/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification/ResourceClassificationInstanceProvider.cs (modified) (2 diffs)

branches/3040_VectorBasedGP/HeuristicLab.Problems.Instances.DataAnalysis/3.3/Classification/ResourceClassificationInstanceProvider.cs

@@ -40 +40 @@
       using (var instancesZipFile = new ZipArchive(GetType().Assembly.GetManifestResourceStream(instanceArchiveName), ZipArchiveMode.Read)) {
         var entry = instancesZipFile.GetEntry(descriptor.ResourceName);
-        NumberFormatInfo numberFormat;
-        DateTimeFormatInfo dateFormat;
-        char separator;
-        using (Stream stream = entry.Open()) {
-          TableFileParser.DetermineFileFormat(stream, out numberFormat, out dateFormat, out separator);
-        }
+
+        var formatOptions = GetFormatOptions(entry);
 
         TableFileParser csvFileParser = new TableFileParser();
         using (Stream stream = entry.Open()) {
-          csvFileParser.Parse(stream, numberFormat, dateFormat, separator, true);
+          csvFileParser.Parse(stream, formatOptions, true);
         }
 
@@ -65 +61 @@
               .Where(x => Regex.Match(x, @".*\.Data\." + fileName).Success).SingleOrDefault();
     }
+
+    protected virtual TableFileFormatOptions GetFormatOptions(ZipArchiveEntry entry) {
+      using (Stream stream = entry.Open()) {
+        return TableFileParser.DetermineFileFormat(stream);
+      }
+    }
   }
 }