//convert number-based list to letter-based list
static void getNewTagList(baseHashMap <int, string> tagMap, ref List <string> tagList)
{
List <string> tmpList = new List <string>();
foreach (string im in tagList)
{
string[] tagAry = im.Split(Global.commaAry, StringSplitOptions.RemoveEmptyEntries);
for (int i = 0; i < tagAry.Length; i++)
{
int index = int.Parse(tagAry[i]);
if (!tagMap.ContainsKey(index))
{
throw new Exception("error");
}
tagAry[i] = tagMap[index];
}
string newTags = string.Join(",", tagAry);
tmpList.Add(newTags);
}
tagList.Clear();
foreach (string im in tmpList)
{
tagList.Add(im);
}
}
//public void getMaps(string file)
//{
// StreamReader swFeat = new StreamReader(Global.modelDir + "/featureIndex.txt");
// string line = "";
// int i = 0;
// while ((line = swFeat.ReadLine()) != null)
// {
// string[] strs = line.Split(' ');
// featureIndexMap[strs[0]] = i;
// i += 1;
// }
// swFeat.Close();
// StreamReader swTag = new StreamReader(Global.modelDir + "/tagIndex.txt");
// line = "";
// i = 0;
// while ((line = swTag.ReadLine()) != null)
// {
// string[] strs = line.Split(' ');
// tagIndexMap[strs[0]] = i;
// i += 1;
// }
// swTag.Close();
//}
//for small memory load, should read line by line
public void convertFile(string file)
{
if (!File.Exists(file))
{
Console.WriteLine("file {0} no exist!", file);
return;
}
//Console.WriteLine("file {0} converting...", file);
Console.WriteLine("file converting...");
StreamReader sr = new StreamReader(file);
//convert to files of new format
StreamWriter swFeature, swGold;
if (file == Global.fTrain)
{
swFeature = new StreamWriter(Global.fFeatureTrain);
swGold = new StreamWriter(Global.fGoldTrain);
}
else
{
swFeature = new StreamWriter(Global.fFeatureTest);
swGold = new StreamWriter(Global.fGoldTest);
}
swFeature.WriteLine(featureIndexMap.Count);
swFeature.WriteLine();
swGold.WriteLine(tagIndexMap.Count);
swGold.WriteLine();
while (!sr.EndOfStream)
{
string line = sr.ReadLine();
line = line.Replace("\t", " ");
line = line.Replace("\r", "\r");
if (line == "")//end of a sample
{
swFeature.WriteLine();
swGold.WriteLine();
swGold.WriteLine();
continue;
}
int flag = 0;
string[] ary = line.Split(Global.blankAry, StringSplitOptions.RemoveEmptyEntries);
for (int i = 1; i < ary.Length - 1; i++)
{
if (ary[i] == "/")//no feature here
{
continue;
}
string[] ary2 = ary[i].Split(Global.slashAry, StringSplitOptions.RemoveEmptyEntries);//for real-value features
string feature = i.ToString() + "." + ary2[0];
string value = "";
bool real = false;
if (ary2.Length > 1)
{
value = ary2[1];
real = true;
}
if (featureIndexMap.ContainsKey(feature) == false)
{
continue;
}
flag = 1;
int fIndex = featureIndexMap[feature];
if (!real)
{
swFeature.Write("{0},", fIndex);
}
else
{
swFeature.Write("{0}/{1},", fIndex, value);
}
}
if (flag == 0)
{
swFeature.Write("0");
}
swFeature.WriteLine();
string tag = ary[ary.Length - 1];
int tIndex = tagIndexMap[tag];
swGold.Write("{0},", tIndex);
}
sr.Close();
swFeature.Close();
swGold.Close();
}