/// <summary>
/// Save feature weights into file
/// </summary>
/// <param name="filename"></param>
/// <param name="bVQ"></param>
/// <returns></returns>
public void SaveFeatureWeight(string filename, bool bVQ)
{
var filename_alpha = filename + ".alpha";
var tofs = new StreamWriter(filename_alpha, false);
var bw = new BinaryWriter(tofs.BaseStream);
if (bVQ == true)
{
Logger.WriteLine("Save feature weights into a VQ model: {0}", filename_alpha);
//Build code book
VectorQuantization vq = new VectorQuantization();
for (long i = 1; i <= maxid_; i++)
{
vq.Add(alpha_[i]);
}
int vqSize = 256;
double distortion = vq.BuildCodebook(vqSize);
Logger.WriteLine("Weight vector quantization distortion: {0}", distortion);
//VQ size
bw.Write(vqSize);
//Save VQ codebook into file
for (int j = 0; j < vqSize; j++)
{
bw.Write(vq.CodeBook[j]);
}
//Save weights
for (long i = 1; i <= maxid_; ++i)
{
bw.Write((byte)vq.ComputeVQ(alpha_[i]));
}
}
else
{
Logger.WriteLine("Save feature weights into a normal model: {0}", filename_alpha);
bw.Write(0);
//Save weights
for (long i = 1; i <= maxid_; ++i)
{
bw.Write((float)alpha_[i]);
}
}
bw.Close();
}
private void SaveLSTMWeights(Vector4[][] weight, BinaryWriter fo, bool bVQ = false)
{
var w = weight.Length;
var h = weight[0].Length;
var vqSize = 256;
Logger.WriteLine("Saving LSTM weight matrix. width:{0}, height:{1}, vq:{2}", w, h, bVQ);
fo.Write(weight.Length);
fo.Write(weight[0].Length);
if (bVQ == false)
{
fo.Write(0);
for (var i = 0; i < w; i++)
{
for (var j = 0; j < h; j++)
{
fo.Write(weight[i][j].X);
fo.Write(weight[i][j].Y);
fo.Write(weight[i][j].Z);
fo.Write(weight[i][j].W);
}
}
}
else
{
//Build vector quantization model
var vqInputCell = new VectorQuantization();
var vqInputForgetGate = new VectorQuantization();
var vqInputInputGate = new VectorQuantization();
var vqInputOutputGate = new VectorQuantization();
for (var i = 0; i < w; i++)
{
for (var j = 0; j < h; j++)
{
vqInputInputGate.Add(weight[i][j].X);
vqInputForgetGate.Add(weight[i][j].Y);
vqInputCell.Add(weight[i][j].Z);
vqInputOutputGate.Add(weight[i][j].W);
}
}
var distortion = vqInputInputGate.BuildCodebook(vqSize);
Logger.WriteLine("InputInputGate distortion: {0}", distortion);
distortion = vqInputForgetGate.BuildCodebook(vqSize);
Logger.WriteLine("InputForgetGate distortion: {0}", distortion);
distortion = vqInputCell.BuildCodebook(vqSize);
Logger.WriteLine("InputCell distortion: {0}", distortion);
distortion = vqInputOutputGate.BuildCodebook(vqSize);
Logger.WriteLine("InputOutputGate distortion: {0}", distortion);
fo.Write(vqSize);
//Save InputInputGate VQ codebook into file
for (var j = 0; j < vqSize; j++)
{
fo.Write(vqInputInputGate.CodeBook[j]);
}
//Save InputForgetGate VQ codebook into file
for (var j = 0; j < vqSize; j++)
{
fo.Write(vqInputForgetGate.CodeBook[j]);
}
//Save InputCell VQ codebook into file
for (var j = 0; j < vqSize; j++)
{
fo.Write(vqInputCell.CodeBook[j]);
}
//Save InputOutputGate VQ codebook into file
for (var j = 0; j < vqSize; j++)
{
fo.Write(vqInputOutputGate.CodeBook[j]);
}
for (var i = 0; i < w; i++)
{
for (var j = 0; j < h; j++)
{
fo.Write((byte)vqInputInputGate.ComputeVQ(weight[i][j].X));
fo.Write((byte)vqInputForgetGate.ComputeVQ(weight[i][j].Y));
fo.Write((byte)vqInputCell.ComputeVQ(weight[i][j].Z));
fo.Write((byte)vqInputOutputGate.ComputeVQ(weight[i][j].W));
}
}
}
}
//Save matrix into file as binary format
public static void SaveMatrix(Matrix <float> mat, BinaryWriter fo, bool bVQ = false)
{
//Save the width and height of the matrix
fo.Write(mat.Width);
fo.Write(mat.Height);
if (bVQ == false)
{
Logger.WriteLine("Saving matrix without VQ...");
fo.Write(0); // non-VQ
//Save the data in matrix
for (var r = 0; r < mat.Height; r++)
{
for (var c = 0; c < mat.Width; c++)
{
fo.Write(mat[r][c]);
}
}
}
else
{
//Build vector quantization matrix
var vqSize = 256;
var vq = new VectorQuantization();
Logger.WriteLine("Saving matrix with VQ {0}...", vqSize);
var valSize = 0;
for (var i = 0; i < mat.Height; i++)
{
for (var j = 0; j < mat.Width; j++)
{
vq.Add(mat[i][j]);
valSize++;
}
}
if (vqSize > valSize)
{
vqSize = valSize;
}
var distortion = vq.BuildCodebook(vqSize);
Logger.WriteLine("Distortion: {0}, vqSize: {1}", distortion, vqSize);
//Save VQ codebook into file
fo.Write(vqSize);
for (var j = 0; j < vqSize; j++)
{
fo.Write(vq.CodeBook[j]);
}
//Save the data in matrix
for (var r = 0; r < mat.Height; r++)
{
for (var c = 0; c < mat.Width; c++)
{
fo.Write((byte)vq.ComputeVQ(mat[r][c]));
}
}
}
}
public bool BuildVQModel(string strFileName)
{
int vqSize = 256;
if (entireTermList == null || entireTermList.Count == 0)
{
return(false);
}
StreamWriter fo = new StreamWriter(strFileName);
BinaryWriter bw = new BinaryWriter(fo.BaseStream);
// Save the word vectors
bw.Write(entireTermList.Count); //Vocabulary size
bw.Write(vectorSize); //Vector size
bw.Write(vqSize); //VQ size
Logger.WriteLine("vocabulary size: {0}, vector size: {1}, vq size: {2}", entireTermList.Count, vectorSize, vqSize);
//Create word and VQ values mapping table
Dictionary <string, List <byte> > vqResult = new Dictionary <string, List <byte> >();
foreach (Term term in entireTermList)
{
vqResult.Add(term.strTerm, new List <byte>());
}
Logger.WriteLine("Dims Distortion:");
for (int i = 0; i < vectorSize; i++)
{
//Generate VQ values for each dimension
VectorQuantization vq = new VectorQuantization();
for (int j = 0; j < entireTermList.Count; j++)
{
vq.Add(entireTermList[j].vector[i]);
}
double distortion = vq.BuildCodebook(vqSize);
Logger.WriteLine("Dim {0}: {1}", i, distortion);
for (int j = 0; j < entireTermList.Count; j++)
{
byte vqValue = (byte)vq.ComputeVQ(entireTermList[j].vector[i]);
vqResult[entireTermList[j].strTerm].Add(vqValue);
}
//Save VQ codebook into model file
for (int j = 0; j < vqSize; j++)
{
bw.Write(vq.CodeBook[j]);
}
}
foreach (KeyValuePair <string, List <byte> > pair in vqResult)
{
if (pair.Value.Count != vectorSize)
{
throw new Exception(String.Format("word {0} has inconsistent vector size: orginial size is {1}, vq size is {2}",
pair.Key, vectorSize, pair.Value.Count));
}
//term string
bw.Write(pair.Key);
//term vector
for (int b = 0; b < pair.Value.Count; b++)
{
bw.Write(pair.Value[b]);
}
}
bw.Flush();
fo.Flush();
fo.Close();
return(true);
}
//Save matrix into file as binary format
public static void SaveMatrix(Matrix<double> mat, BinaryWriter fo, bool bVQ = false)
{
//Save the width and height of the matrix
fo.Write(mat.Width);
fo.Write(mat.Height);
if (bVQ == false)
{
Logger.WriteLine("Saving matrix without VQ...");
fo.Write(0); // non-VQ
//Save the data in matrix
for (int r = 0; r < mat.Height; r++)
{
for (int c = 0; c < mat.Width; c++)
{
fo.Write(mat[r][c]);
}
}
}
else
{
//Build vector quantization matrix
int vqSize = 256;
VectorQuantization vq = new VectorQuantization();
Logger.WriteLine("Saving matrix with VQ {0}...", vqSize);
int valSize = 0;
for (int i = 0; i < mat.Height; i++)
{
for (int j = 0; j < mat.Width; j++)
{
vq.Add(mat[i][j]);
valSize++;
}
}
if (vqSize > valSize)
{
vqSize = valSize;
}
double distortion = vq.BuildCodebook(vqSize);
Logger.WriteLine("Distortion: {0}, vqSize: {1}", distortion, vqSize);
//Save VQ codebook into file
fo.Write(vqSize);
for (int j = 0; j < vqSize; j++)
{
fo.Write(vq.CodeBook[j]);
}
//Save the data in matrix
for (int r = 0; r < mat.Height; r++)
{
for (int c = 0; c < mat.Width; c++)
{
fo.Write((byte)vq.ComputeVQ(mat[r][c]));
}
}
}
}
private void saveLSTMWeight(Vector4[][] weight, BinaryWriter fo, bool bVQ = false)
{
int w = weight.Length;
int h = weight[0].Length;
int vqSize = 256;
Logger.WriteLine("Saving LSTM weight matrix. width:{0}, height:{1}, vq:{2}", w, h, bVQ);
fo.Write(weight.Length);
fo.Write(weight[0].Length);
if (bVQ == false)
{
fo.Write(0);
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
fo.Write(weight[i][j].X);
fo.Write(weight[i][j].Y);
fo.Write(weight[i][j].Z);
fo.Write(weight[i][j].W);
}
}
}
else
{
//Build vector quantization model
VectorQuantization vqInputCell = new VectorQuantization();
VectorQuantization vqInputForgetGate = new VectorQuantization();
VectorQuantization vqInputInputGate = new VectorQuantization();
VectorQuantization vqInputOutputGate = new VectorQuantization();
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
vqInputInputGate.Add(weight[i][j].X);
vqInputForgetGate.Add(weight[i][j].Y);
vqInputCell.Add(weight[i][j].Z);
vqInputOutputGate.Add(weight[i][j].W);
}
}
double distortion = 0.0;
distortion = vqInputInputGate.BuildCodebook(vqSize);
Logger.WriteLine("InputInputGate distortion: {0}", distortion);
distortion = vqInputForgetGate.BuildCodebook(vqSize);
Logger.WriteLine("InputForgetGate distortion: {0}", distortion);
distortion = vqInputCell.BuildCodebook(vqSize);
Logger.WriteLine("InputCell distortion: {0}", distortion);
distortion = vqInputOutputGate.BuildCodebook(vqSize);
Logger.WriteLine("InputOutputGate distortion: {0}", distortion);
fo.Write(vqSize);
//Save InputInputGate VQ codebook into file
for (int j = 0; j < vqSize; j++)
{
fo.Write(vqInputInputGate.CodeBook[j]);
}
//Save InputForgetGate VQ codebook into file
for (int j = 0; j < vqSize; j++)
{
fo.Write(vqInputForgetGate.CodeBook[j]);
}
//Save InputCell VQ codebook into file
for (int j = 0; j < vqSize; j++)
{
fo.Write(vqInputCell.CodeBook[j]);
}
//Save InputOutputGate VQ codebook into file
for (int j = 0; j < vqSize; j++)
{
fo.Write(vqInputOutputGate.CodeBook[j]);
}
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
fo.Write((byte)vqInputInputGate.ComputeVQ(weight[i][j].X));
fo.Write((byte)vqInputForgetGate.ComputeVQ(weight[i][j].Y));
fo.Write((byte)vqInputCell.ComputeVQ(weight[i][j].Z));
fo.Write((byte)vqInputOutputGate.ComputeVQ(weight[i][j].W));
}
}
}
}
private void saveLSTMWeight(LSTMWeight[][] weight, BinaryWriter fo)
{
int w = weight.Length;
int h = weight[0].Length;
int vqSize = 256;
Logger.WriteLine("Saving LSTM weight matrix. width:{0}, height:{1}, vqSize:{2}", w, h, vqSize);
fo.Write(weight.Length);
fo.Write(weight[0].Length);
//Build vector quantization model
VectorQuantization vq = new VectorQuantization();
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
vq.Add(weight[i][j].wInputCell);
vq.Add(weight[i][j].wInputForgetGate);
vq.Add(weight[i][j].wInputInputGate);
vq.Add(weight[i][j].wInputOutputGate);
}
}
double distortion = vq.BuildCodebook(vqSize);
Logger.WriteLine("Distortion: {0}", distortion);
//Save VQ codebook into file
fo.Write(vqSize);
for (int j = 0; j < vqSize; j++)
{
fo.Write(vq.CodeBook[j]);
}
for (int i = 0; i < w; i++)
{
for (int j = 0; j < h; j++)
{
fo.Write((byte)vq.ComputeVQ(weight[i][j].wInputCell));
fo.Write((byte)vq.ComputeVQ(weight[i][j].wInputForgetGate));
fo.Write((byte)vq.ComputeVQ(weight[i][j].wInputInputGate));
fo.Write((byte)vq.ComputeVQ(weight[i][j].wInputOutputGate));
}
}
}