public static string testAllLanguages(LangDescriptor[] languages, string[] corpusDirs, string imageFileName)
{
IList <string> languageNames = BuffUtils.map(languages, l => l.name + "_err");
// Collections.sort(languageNames);
IDictionary <string, int?> corpusSizes = new Dictionary <string, int?>();
for (int i = 0; i < languages.Length; i++)
{
LangDescriptor language = languages[i];
IList <string> filenames = Tool.getFilenames(corpusDirs[i], language.fileRegex);
corpusSizes[language.name] = filenames.Count;
}
IList <string> languageNamesAsStr = BuffUtils.map(languages, l => '"' + l.name + "\\nn=" + corpusSizes[l.name] + '"');
// Collections.sort(languageNamesAsStr);
StringBuilder data = new StringBuilder();
for (int i = 0; i < languages.Length; i++)
{
LangDescriptor language = languages[i];
string corpus = corpusDirs[i];
LeaveOneOutValidator validator = new LeaveOneOutValidator(corpus, language);
Triple <IList <Formatter>, IList <float>, IList <float> > results = validator.validateDocuments(true, "/tmp");
IList <Formatter> formatters = results.a;
IList <float> distances = results.b;
IList <float> errors = results.c;
// data.append(language.name+"_dist = "+distances+"\n");
data.Append(language.name + "_err = " + errors + "\n");
}
string python = "#\n" + "# AUTO-GENERATED FILE. DO NOT EDIT\n" + "# CodeBuff %s '%s'\n" + "#\n" + "import numpy as np\n" + "import pylab\n" + "import matplotlib.pyplot as plt\n\n" + "%s\n" + "language_data = %s\n" + "labels = %s\n" + "fig = plt.figure()\n" + "ax = plt.subplot(111)\n" + "ax.boxplot(language_data,\n" + " whis=[10, 90], # 10 and 90 %% whiskers\n" + " widths=.35,\n" + " labels=labels,\n" + " showfliers=False)\n" + "ax.set_xticklabels(labels, rotation=60, fontsize=18)\n" + "ax.tick_params(axis='both', which='major', labelsize=18)\n" + "plt.xticks(range(1,len(labels)+1), labels, rotation=60, fontsize=18)\n" + "pylab.ylim([0,.28])\n" + "ax.yaxis.grid(True, linestyle='-', which='major', color='lightgrey', alpha=0.5)\n" + "ax.set_xlabel(\"Grammar and corpus size\", fontsize=20)\n" + "ax.set_ylabel(\"Misclassification Error Rate\", fontsize=20)\n" + "# ax.set_title(\"Leave-one-out Validation Using Error Rate\\nBetween Formatted and Original File\")\n" + "plt.tight_layout()\n" + "fig.savefig('images/%s', format='pdf')\n" + "plt.show()\n";
return(string.Format(python, Tool.version, DateTime.Now, data, languageNames, languageNamesAsStr, imageFileName));
}
public static IList <float?> checkStability(LangDescriptor language)
{
IList <float?> errorRates = new List <float?>();
// format the corpus into tmp dir
LeaveOneOutValidator validator0 = new LeaveOneOutValidator(language.corpusDir, language);
Triple <IList <Formatter>, IList <float>, IList <float> > results0 = validator0.validateDocuments(false, "/tmp/stability/1");
errorRates.Add(BuffUtils.median(results0.c));
IList <Formatter> formatters0 = results0.a;
// now try formatting it over and over
for (int i = 1; i <= STAGES; i++)
{
string inputDir = "/tmp/stability/" + i;
string outputDir = "/tmp/stability/" + (i + 1);
LeaveOneOutValidator validator = new LeaveOneOutValidator(inputDir, language);
Triple <IList <Formatter>, IList <float>, IList <float> > results = validator.validateDocuments(false, outputDir);
IList <Formatter> formatters = results.a;
IList <float?> distances = new List <float?>();
for (int j = 0; j < formatters.Count; j++)
{
Formatter f0 = formatters0[j];
Formatter f = formatters[j];
float editDistance = Dbg.normalizedLevenshteinDistance(f.Output, f0.Output);
distances.Add(editDistance);
}
errorRates.Add(BuffUtils.median(distances));
}
return(errorRates);
}
//JAVA TO C# CONVERTER WARNING: Method 'throws' clauses are not available in .NET:
//ORIGINAL LINE: public static void main(String[] args) throws Exception
public static void Main(string[] args)
{
LeaveOneOutValidator validator = new LeaveOneOutValidator(JAVA8_DESCR.corpusDir, JAVA8_DESCR);
Triple <IList <Formatter>, IList <float?>, IList <float?> > results = validator.validateDocuments(false, "output");
Log.WriteLine(results.b);
Log.WriteLine(results.c);
}
public static void Main(string[] args)
{
LeaveOneOutValidator validator = new LeaveOneOutValidator(Tool.ANTLR4_DESCR.corpusDir, Tool.ANTLR4_DESCR);
Triple <IList <Formatter>, IList <float>, IList <float> > results = validator.validateDocuments(false, "output");
Console.WriteLine(results.b);
Console.WriteLine(results.c);
}
public static void Main(string[] args)
{
LangDescriptor[] languages = new LangDescriptor[] { JAVA_DESCR, JAVA8_DESCR, JAVA_GUAVA_DESCR };
IList <string> corpusDirs = BuffUtils.map(languages, l => l.corpusDir);
string[] dirs = corpusDirs.ToArray();
string python = LeaveOneOutValidator.testAllLanguages(languages, dirs, "all_java_leave_one_out.pdf");
string fileName = "python/src/all_java_leave_one_out.py";
org.antlr.codebuff.misc.Utils.writeFile(fileName, python);
Log.WriteLine("wrote python code to " + fileName);
}
public static void Main(string[] args)
{
LangDescriptor[] languages = new LangDescriptor[] { QUORUM_DESCR, JAVA_DESCR, JAVA8_DESCR, JAVA_GUAVA_DESCR, ANTLR4_DESCR, SQLITE_CLEAN_DESCR, TSQL_CLEAN_DESCR };
// walk and generator output but no edit distance
for (int i = 0; i < languages.Length; i++)
{
LangDescriptor language = languages[i];
LeaveOneOutValidator validator = new LeaveOneOutValidator(language.corpusDir, language);
validator.validateDocuments(false, "output");
}
}
public static IList <float> getAlignmentErrorRates(LangDescriptor language, FeatureMetaData[] injectWSFeatures, FeatureMetaData[] alignmentFeatures)
{
LeaveOneOutValidator validator = new LeaveOneOutValidator(language.corpusDir, language);
Triple <IList <Formatter>, IList <float>, IList <float> > results = validator.validateDocuments(injectWSFeatures, alignmentFeatures, false, null);
IList <Formatter> formatters = results.a;
IList <float> alignErrorRates = new List <float>(); // don't include align errors
foreach (Formatter formatter in formatters)
{
ClassificationAnalysis analysis = new ClassificationAnalysis(formatter.testDoc, formatter.AnalysisPerToken);
alignErrorRates.Add(analysis.AlignmentErrorRate);
}
// System.out.println(results.c);
// System.out.println("vs");
// System.out.println(alignErrorRates);
return(alignErrorRates);
}
public static void Main(string[] args)
{
// we need to get all of the results in order so that we can compare
LeaveOneOutValidator.FORCE_SINGLE_THREADED = true;
float sql_median;
float java_st_median;
float java_guava_median;
{
// JAVA
IList <float?> distances = new List <float?>();
LeaveOneOutValidator javaValidator = new LeaveOneOutValidator(Tool.JAVA_DESCR.corpusDir, JAVA_DESCR);
LeaveOneOutValidator java8Validator = new LeaveOneOutValidator(JAVA8_DESCR.corpusDir, JAVA8_DESCR);
Triple <IList <Formatter>, IList <float?>, IList <float?> > javaResults = javaValidator.validateDocuments(false, null);
Triple <IList <Formatter>, IList <float?>, IList <float?> > java8Results = java8Validator.validateDocuments(false, null);
IList <Formatter> javaFormatters = javaResults.a;
IList <Formatter> java8Formatters = java8Results.a;
for (int i = 0; i < javaFormatters.Count; i++)
{
Formatter java = javaFormatters[i];
Formatter java8 = java8Formatters[i];
float editDistance = Dbg.normalizedLevenshteinDistance(java.Output, java8.Output);
distances.Add(editDistance);
// System.out.println(java.testDoc.fileName+" edit distance "+editDistance);
}
{
distances.Sort();
int n = distances.Count;
float min = distances[0].Value;
float quart = distances[(int)(0.27 * n)].Value;
float median = distances[n / 2].Value;
float quart3 = distances[(int)(0.75 * n)].Value;
float max = distances[distances.Count - 1].Value;
string display = "(" + min + "," + median + "," + max + ")";
java_st_median = median;
}
}
{
// JAVA GUAVA
IList <float?> distances = new List <float?>();
LeaveOneOutValidator java_guavaValidator = new LeaveOneOutValidator(JAVA_GUAVA_DESCR.corpusDir, JAVA_GUAVA_DESCR);
LeaveOneOutValidator java8_guavaValidator = new LeaveOneOutValidator(JAVA8_GUAVA_DESCR.corpusDir, JAVA8_GUAVA_DESCR);
Triple <IList <Formatter>, IList <float?>, IList <float?> > java_guavaResults = java_guavaValidator.validateDocuments(false, null);
Triple <IList <Formatter>, IList <float?>, IList <float?> > java8_guavaResults = java8_guavaValidator.validateDocuments(false, null);
IList <Formatter> java_guavaFormatters = java_guavaResults.a;
IList <Formatter> java8_guavaFormatters = java8_guavaResults.a;
for (int i = 0; i < java_guavaFormatters.Count; i++)
{
Formatter java_guava = java_guavaFormatters[i];
Formatter java8_guava = java8_guavaFormatters[i];
float editDistance = Dbg.normalizedLevenshteinDistance(java_guava.Output, java8_guava.Output);
distances.Add(editDistance);
// System.out.println(java_guava.testDoc.fileName+" edit distance "+editDistance);
}
{
distances.Sort();
int n = distances.Count;
float min = distances[0].Value;
float quart = distances[(int)(0.27 * n)].Value;
float median = distances[n / 2].Value;
float quart3 = distances[(int)(0.75 * n)].Value;
float max = distances[distances.Count - 1].Value;
string display = "(" + min + "," + median + "," + max + ")";
java_guava_median = median;
}
}
Console.WriteLine("clean SQLite vs TSQL edit distance info median=" + sql_median);
Console.WriteLine("Java vs Java8 edit distance info median=" + java_st_median);
Console.WriteLine("Java vs Java8 guava edit distance info median=" + java_guava_median);
}