static void Main(string[] args)
{
Console.WriteLine("Choose operation by writing the corresponding choice (1 or 2):");
Console.WriteLine("(1) Parse Retrovirus Database Html Dump And Generate Positions File");
Console.WriteLine("(2) Generate Training and Validation Set From Existing Positions File");
string optionChosen = Console.ReadLine();
while (!optionChosen.Contains("1") && !optionChosen.Contains("2"))
{
Console.WriteLine("Unknown option selected, please choose (1 or 2)!");
Console.WriteLine("(1) Parse Retrovirus Database Html Dump And Generate Positions File");
Console.WriteLine("(2) Generate Training and Validation Set From Existing Positions File");
optionChosen = Console.ReadLine();
}
string positionsPath = workingDirectoryFilePath + "positions.txt";
string labelsPath = workingDirectoryFilePath + "labels.txt";
string insertsPath = workingDirectoryFilePath + "inserts.txt";
string DNAFileOutPath = workingDirectoryFilePath + "DNAString.txt";
string DNAFileOutValidationPath = workingDirectoryFilePath + "DNAString_validation.txt";
string validationInsertsPath = workingDirectoryFilePath + "inserts_validation.txt";
string validationLabelsPath = workingDirectoryFilePath + "labels_validation.txt";
if (optionChosen.Contains("1"))
{
Console.WriteLine("Make sure that you have the dump (result.html) in same path as this executable.");
Console.WriteLine("Press ENTER to continue parsing the dump and generating the positions file");
Console.ReadLine();
Parser.generatePositionsFile(positionsPath, new System.Collections.Hashtable());
Console.WriteLine("DONE!, Press ENTER to exit");
Console.ReadLine();
}
else if (optionChosen.Contains("2"))
{
Console.WriteLine("Make sure that you have the positions file (positions.txt) in same path as this executable.");
Console.WriteLine("Press ENTER to continue parsing the dump and generating the positions file");
Console.ReadLine();
Console.WriteLine("Reading existing positions...");
System.Collections.Hashtable existingPositionsHash = DatasetGeneratorUtil.getExistingPositionsSorted(positionsPath);
Console.WriteLine("Done reading positions, generating data..");
DatasetFileGenerator.getExistingPositionsAndGenerateData(positionsPath, labelsPath, insertsPath, validationInsertsPath, validationLabelsPath, DNAFileOutPath, DNAFileOutValidationPath, existingPositionsHash);
Console.WriteLine("DONE!, Press ENTER to exit");
Console.ReadLine();
}
}
public static async void getExistingPositionsAndGenerateData(string positionsFile, string labelsFileOutputPath,
string insertsFileOutputPath, string validationFileOutputPath, string validationLabelsFileOutputPath,
string DNAStringOutputFilePath, string DNAStringOutputValidationFilePath, System.Collections.Hashtable existingPositionsHash)
{
float totalTargetLength;
float percentDone;
List <int> existingPositions = new List <int>();
string tmp;
string[] splitted;
int position;
Task <DNATuple> tskTup1;
Task <DNATuple> tskTup2;
DNATuple tup;
bool printToValidation = false;
KeyValuePair <string, int> randomFalsePosition;
using (FileStream fs = File.Open(positionsFile, FileMode.Open, FileAccess.Read, FileShare.ReadWrite))
using (BufferedStream bs = new BufferedStream(fs))
using (StreamReader sr = new StreamReader(bs))
{
using (StreamWriter swValidation = File.AppendText(validationFileOutputPath))
{
using (StreamWriter swValidationLabels = File.AppendText(validationLabelsFileOutputPath))
{
using (StreamWriter swLabels = File.AppendText(labelsFileOutputPath))
{
using (StreamWriter swInserts = File.AppendText(insertsFileOutputPath))
{
using (StreamWriter swDNAOutput = File.AppendText(DNAStringOutputFilePath))
{
using (StreamWriter swDNAValidationOutput = File.AppendText(DNAStringOutputValidationFilePath))
{
totalTargetLength = sr.BaseStream.Length;
while (sr.Peek() != -1)
{
tmp = sr.ReadLine();
splitted = tmp.Split(',');
position = Int32.Parse(splitted[0]);
tskTup1 = Task <DNATuple> .Factory.StartNew(() => DatasetGeneratorUtil.getDNASequence(position, splitted[1].Replace("_", "")));
randomFalsePosition = DatasetGeneratorUtil.getRandomPostion(existingPositionsHash);
tskTup2 = Task <DNATuple> .Factory.StartNew(() => DatasetGeneratorUtil.getDNASequence(randomFalsePosition.Value, randomFalsePosition.Key));
tskTup1.Wait();
tup = tskTup1.Result;
if (tup == null)
{
continue; //unknown chromosome file specified, continue
}
tmp = DatasetGeneratorUtil.DNAStringToOneHotEncoding(tup.beforePosition + tup.afterPosition);
if (printToValidation)
{
swDNAValidationOutput.WriteLine(tup.beforePosition + tup.afterPosition);
swValidation.WriteLine(tmp);
swValidationLabels.WriteLine("1");
}
else
{
swDNAOutput.WriteLine(tup.beforePosition + tup.afterPosition);
swInserts.WriteLine(tmp);
swLabels.WriteLine("1");
}
tskTup2.Wait();
tup = tskTup2.Result;
tmp = DatasetGeneratorUtil.DNAStringToOneHotEncoding(tup.beforePosition + tup.afterPosition);
if (printToValidation)
{
swDNAValidationOutput.WriteLine(tup.beforePosition + tup.afterPosition);
swValidation.WriteLine(tmp);
swValidationLabels.WriteLine("0");
}
else
{
swDNAOutput.WriteLine(tup.beforePosition + tup.afterPosition);
swInserts.WriteLine(tmp);
swLabels.WriteLine("0");
}
percentDone = ((float)sr.BaseStream.Position) / totalTargetLength;
if (percentDone > 0.90f)
{
printToValidation = true;
}
Console.WriteLine("Creating files:" + percentDone.ToString("0.0000") + "%");
}
swInserts.Flush();
swLabels.Flush();
swValidation.Flush();
}
}
}
}
}
}
}
}
