hanzoai/jube-fork
2
0
Fork 0
mirror of https://github.com/hanzoai/jube-fork.git synced 2026-10-30 18:33:33 +00:00
Code Issues Packages Projects Releases Wiki Activity
Files
118-cache-use-reduction-features
jube-fork/Accord.MachineLearning/Clustering/TSNE.cs
T
richard-churchman cc0a2bee21 Extensive modifications to the serialisation of Neural Network models to support Newtonsoft Json.net rather than Binary Serializer. 
Included open-source Accord.net library code natively to the solution given the deprecated state of that project.  Accord.net will be supported inside Jube for own needs (such as upgrading to .Net 7+).

Accord.net libraries modified to compile to .Net 6.

Modified Accord.net libraries to better support Newtonsoft Json.net serialisation.

Included annotations in the Accord.net libraries to house important dependency statistics. Serialised json can be used independently of Jube to recall a Neural Network.

Removal of Binary Serializer totally as a precursor to the projects support of .Net 7+.

Extensive bug fixing in crucial functions of Neural Network training and recall.  While this part of the system remains in Alpha, it is substantially more fit for purpose.
2023-12-24 10:22:22 +01:00

874 lines
33 KiB
C#
Raw Permalink Blame History

// Accord Machine Learning Library
// The Accord.NET Framework
// http://accord-framework.net
//
// Copyright © César Souza, 2009-2017
// cesarsouza at gmail.com
//
// This library is free software; you can redistribute it and/or
// modify it under the terms of the GNU Lesser General Public
// License as published by the Free Software Foundation; either
// version 2.1 of the License, or (at your option) any later version.
//
// This library is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
// Lesser General Public License for more details.
//
// You should have received a copy of the GNU Lesser General Public
// License along with this library; if not, write to the Free Software
// Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
//
// The code contained in this class was adapted from Laurens van der Maaten excellent
// BH T-SNE code from https://github.com/lvdmaaten/bhtsne/. The original license is
// listed below:
//
// Copyright (c) 2014, Laurens van der Maaten (Delft University of Technology)
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
// 1. Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
// 3. All advertising materials mentioning features or use of this software
// must display the following acknowledgement:
// This product includes software developed by the Delft University of Technology.
// 4. Neither the name of the Delft University of Technology nor the names of
// its contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY LAURENS VAN DER MAATEN ''AS IS'' AND ANY EXPRESS
// OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
// OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
// EVENT SHALL LAURENS VAN DER MAATEN BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
// BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
// CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
// IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
// OF SUCH DAMAGE.
//
namespace Accord.MachineLearning.Clustering
{
#if !MONO
using Accord.Collections;
using Accord.Statistics.Distributions.Univariate;
using System;
using System.Collections.Generic;
using System.Diagnostics;
using System.IO;
using System.Linq;
using Accord.Math;
using Accord.Compat;
/// <summary>
/// Barnes-Hutt t-SNE.
/// </summary>
///
/// <remarks>
/// <para>
/// The code contained in this class was adapted from Laurens van der Maaten excellent
/// BH T-SNE code from https://github.com/lvdmaaten/bhtsne/. The original license is
/// listed below:</para>
/// <code>
/// Copyright (c) 2014, Laurens van der Maaten (Delft University of Technology)
/// All rights reserved.
///
/// Redistribution and use in source and binary forms, with or without
/// modification, are permitted provided that the following conditions are met:
/// 1. Redistributions of source code must retain the above copyright
/// notice, this list of conditions and the following disclaimer.
/// 2. Redistributions in binary form must reproduce the above copyright
/// notice, this list of conditions and the following disclaimer in the
/// documentation and/or other materials provided with the distribution.
/// 3. All advertising materials mentioning features or use of this software
/// must display the following acknowledgement:
/// This product includes software developed by the Delft University of Technology.
/// 4. Neither the name of the Delft University of Technology nor the names of
/// its contributors may be used to endorse or promote products derived from
/// this software without specific prior written permission.
///
/// THIS SOFTWARE IS PROVIDED BY LAURENS VAN DER MAATEN ''AS IS'' AND ANY EXPRESS
/// OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
/// OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO
/// EVENT SHALL LAURENS VAN DER MAATEN BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
/// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
/// PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR
/// BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
/// CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
/// IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY
/// OF SUCH DAMAGE.
/// </code>
/// </remarks>
///
/// <example>
/// <code source="Unit Tests\Accord.Tests.MachineLearning\Clustering\TSNETest.cs" region="doc_learn" />
/// </example>
///
public class TSNE : MultipleTransformBase<double[], double>
{
const double DBL_MAX = -1.7976931348623157e+308;
const double DBL_MIN = 2.2250738585072014e-308;
const double FLT_MIN = 1.175494351e-38F;
double perplexity = 50;
double theta = 0.5;
/// <summary>
/// Initializes a new instance of the <see cref="TSNE"/> class.
/// </summary>
///
public TSNE()
{
NumberOfOutputs = 2;
}
/// <summary>
/// Gets or sets t-SNE's perplexity value. Default is 50.
/// </summary>
///
public double Perplexity
{
get { return perplexity; }
set { perplexity = value; }
}
/// <summary>
/// Gets or sets t-SNE's Theta value. Default is 0.5
/// </summary>
///
public double Theta
{
get { return theta; }
set { theta = value; }
}
/// <summary>
/// Not supported.
/// </summary>
///
public override double[] Transform(double[] input, double[] result)
{
return Transform(new[] { input }, new[] { result })[0];
}
/// <summary>
/// Applies the transformation to an input, producing an associated output.
/// </summary>
///
/// <param name="input">The input data to which the transformation should be applied.</param>
/// <param name="result">The location to where to store the result of this transformation.</param>
/// <returns>
/// The output generated by applying this transformation to the given input.
/// </returns>
///
public override double[][] Transform(double[][] input, double[][] result)
{
run(input, result, perplexity, theta);
return result;
}
internal static void run(double[][] X, double[][] Y, double perplexity, double theta, bool skip_random_init = false)
{
int N = X.Rows();
int D = X.Columns();
int no_dims = Y.Columns();
// Determine whether we are using an exact algorithm
if (N - 1 < 3 * perplexity)
throw new Exception(String.Format("Perplexity too large for the number of data points. For {0} points, should be less than {1}", N, (N - 1) / 3.0));
Debug.Write(String.Format("Using no_dims = {0}, perplexity = {1}, and theta = {2}", no_dims, perplexity, theta));
bool exact = (theta == 0.0);
// Set learning parameters
TimeSpan total_time = TimeSpan.Zero;
Stopwatch start;
TimeSpan end;
int max_iter = 1000;
int stop_lying_iter = 250;
int mom_switch_iter = 250;
double momentum = 0.5;
double final_momentum = 0.8;
double eta = 200.0;
// Allocate some memory
double[][] dY = Jagged.Create<double>(N, no_dims);
double[][] uY = Jagged.Create<double>(N, no_dims);
double[][] gains = Jagged.Ones<double>(N, no_dims);
// Normalize input data (to prevent numerical problems)
Debug.Write("Computing input similarities...");
start = Stopwatch.StartNew();
Accord.Statistics.Tools.Center(X, inPlace: true);
X.Divide(X.Max(), result: X);
// Compute input similarities for exact t-SNE
double[][] P = null;
int[] row_P = null;
int[] col_P = null;
double[] val_P = null;
if (exact)
{
Trace.Write("Exact?");
// Compute similarities
P = Jagged.Create<double>(N, N);
computeGaussianPerplexity(X, N, D, ref P, perplexity);
// Symmetrize input similarities
Debug.Write("Symmetrizing...");
for (int n = 0; n < N; n++)
{
for (int m = n + 1; m < N; m++)
{
P[n][m] += P[m][n];
P[m][n] = P[n][m];
}
}
P.Divide(P.Sum(), result: P);
}
// Compute input similarities for approximate t-SNE
else
{
// Compute asymmetric pairwise input similarities
computeGaussianPerplexity(X, N, D, ref row_P, ref col_P, ref val_P, perplexity, (int)(3 * perplexity));
// Symmetrize input similarities
symmetrizeMatrix(ref row_P, ref col_P, ref val_P, N);
double sum_P = 0.0;
for (int i = 0; i < row_P[N]; i++)
sum_P += val_P[i];
for (int i = 0; i < row_P[N]; i++)
val_P[i] /= sum_P;
}
end = start.Elapsed;
// Lie about the P-values
if (exact)
{
P.Multiply(12.0, result: P);
}
else
{
for (int i = 0; i < row_P[N]; i++)
val_P[i] *= 12.0;
}
if (!skip_random_init)
{
// Initialize solution (randomly)
for (int i = 0; i < Y.Length; i++)
for (int j = 0; j < Y[i].Length; j++)
Y[i][j] = randn() * 0.0001;
}
// Perform main training loop
if (exact)
{
Debug.Write(String.Format("Input similarities computed in {0} seconds!", end));
Debug.Write("Learning embedding...");
}
else
{
Debug.Write(String.Format("Input similarities computed in {0} seconds (sparsity = {1})!", end, (double)row_P[N] / ((double)N * (double)N)));
Debug.Write("Learning embedding...");
}
start = Stopwatch.StartNew();
for (int iter = 0; iter < max_iter; iter++)
{
// Compute (approximate) gradient
if (exact)
computeExactGradient(P, Y, N, no_dims, dY);
else
computeGradient(P, row_P, col_P, val_P, Y, N, no_dims, dY, theta);
// Update gains
for (int i = 0; i < gains.Length; i++)
for (int j = 0; j < gains[i].Length; j++)
gains[i][j] = (System.Math.Sign(dY[i][j]) != System.Math.Sign(uY[i][j])) ? (gains[i][j] + 0.2) : (gains[i][j] * 0.8);
for (int i = 0; i < gains.Length; i++)
for (int j = 0; j < gains[i].Length; j++)
if (gains[i][j] < 0.01)
gains[i][j] = 0.01;
// Perform gradient update (with momentum and gains)
for (int i = 0; i < uY.Length; i++)
for (int j = 0; j < uY[i].Length; j++)
uY[i][j] = momentum * uY[i][j] - eta * gains[i][j] * dY[i][j];
for (int i = 0; i < Y.Length; i++)
for (int j = 0; j < Y[i].Length; j++)
Y[i][j] = Y[i][j] + uY[i][j];
// Make solution zero-mean
Accord.Statistics.Tools.Center(Y, inPlace: true);
// Stop lying about the P-values after a while, and switch momentum
if (iter == stop_lying_iter)
{
if (exact)
{
P.Divide(12.0, result: P);
}
else
{
for (int i = 0; i < row_P[N]; i++)
val_P[i] /= 12.0;
}
}
if (iter == mom_switch_iter)
momentum = final_momentum;
// Print out progress
if (iter > 0 && (iter % 50 == 0 || iter == max_iter - 1))
{
end = start.Elapsed;
double C = 0.0;
if (exact)
{
C = evaluateError(P, Y, N, no_dims);
}
else
{
C = evaluateError(row_P, col_P, val_P, Y, N, no_dims, theta); // doing approximate computation here!
}
if (iter == 0)
{
Debug.WriteLine(String.Format("Iteration {0}: error is {1}", iter + 1, C));
}
else
{
total_time += end;
Debug.WriteLine(String.Format("Iteration {0}: error is {1} (50 iterations in {2} seconds)", iter, C, end));
}
start = Stopwatch.StartNew();
}
}
end = start.Elapsed;
total_time += end;
Debug.WriteLine(String.Format("Fitting performed in {0} seconds.", total_time));
}
// Compute gradient of the t-SNE cost function (using Barnes-Hut algorithm)
internal static void computeGradient(double[][] P, int[] inp_row_P, int[] inp_col_P, double[] inp_val_P, double[][] Y, int N, int D, double[][] dC, double theta)
{
// Construct space-partitioning tree on current map
var tree = SPTree.FromData(Y);
// Compute all terms required for t-SNE gradient
var pos_f = Jagged.Create<double>(N, D);
var neg_f = Jagged.Create<double>(N, D);
tree.ComputeEdgeForces(Y, inp_row_P, inp_col_P, inp_val_P, pos_f);
double sum_Q = 0.0;
for (int n = 0; n < Y.Length; n++)
tree.ComputeNonEdgeForces(Y[n], theta, neg_f[n], ref sum_Q);
// Compute final t-SNE gradient
for (int i = 0; i < dC.Length; i++)
for (int j = 0; j < dC[i].Length; j++)
dC[i][j] = pos_f[i][j] - (neg_f[i][j] / sum_Q);
}
// Compute gradient of the t-SNE cost function (exact)
internal static void computeExactGradient(double[][] P, double[][] Y, int N, int D, double[][] dC)
{
// Make sure the current gradient contains zeros
for (int i = 0; i < dC.Length; i++)
for (int j = 0; j < dC[i].Length; j++)
dC[i][j] = 0.0;
// Compute the squared Euclidean distance matrix
double[][] DD = Jagged.Create<double>(N, N);
computeSquaredEuclideanDistance(Y, N, D, DD);
// Compute Q-matrix and normalization sum
double[][] Q = Jagged.Zeros(N, N);
double sum_Q = 0.0;
for (int n = 0; n < N; n++)
{
for (int m = 0; m < N; m++)
{
if (n != m)
{
Q[n][m] = 1.0 / (1.0 + DD[n][m]);
sum_Q += Q[n][m];
}
}
}
// Perform the computation of the gradient
for (int n = 0; n < N; n++)
{
for (int m = 0; m < N; m++)
{
if (n != m)
{
double mult = (P[n][m] - (Q[n][m] / sum_Q)) * Q[n][m];
for (int d = 0; d < D; d++)
dC[n][d] += (Y[n][d] - Y[m][d]) * mult;
}
}
}
}
// Evaluate t-SNE cost function (exactly)
internal static double evaluateError(double[][] P, double[][] Y, int N, int D)
{
// Compute the squared Euclidean distance matrix
double[][] DD = Jagged.Zeros(N, N);
double[][] Q = Jagged.Zeros(N, N);
computeSquaredEuclideanDistance(Y, N, D, DD);
// Compute Q-matrix and normalization sum
double sum_Q = DBL_MIN;
for (int n = 0; n < N; n++)
{
for (int m = 0; m < N; m++)
{
if (n != m)
{
Q[n][m] = 1.0 / (1.0 + DD[n][m]);
sum_Q += Q[n][m];
}
else
{
Q[n][m] = DBL_MIN;
}
}
}
Q.Divide(sum_Q, result: Q);
// Sum t-SNE error
double C = 0.0;
for (int n = 0; n < P.Length; n++)
for (int j = 0; j < P[n].Length; j++)
C += P[n][j] * System.Math.Log((P[n][j] + FLT_MIN) / (Q[n][j] + FLT_MIN));
return C;
}
// Evaluate t-SNE cost function (approximately)
internal static double evaluateError(int[] row_P, int[] col_P, double[] val_P, double[][] Y, int N, int D, double theta)
{
// Get estimate of normalization term
var tree = SPTree.FromData(Y);
double[] buff = new double[D];
double sum_Q = 0.0;
for (int n = 0; n < Y.Length; n++)
tree.ComputeNonEdgeForces(Y[n], theta, buff, ref sum_Q);
// Loop over all edges to compute t-SNE error
double C = 0.0;
for (int n = 0; n < N; n++)
{
for (int i = row_P[n]; i < row_P[n + 1]; i++)
{
int j = col_P[i];
for (int d = 0; d < buff.Length; d++)
buff[d] = Y[n][d];
for (int d = 0; d < buff.Length; d++)
buff[d] -= Y[j][d];
double Q = 0.0;
for (int d = 0; d < buff.Length; d++)
Q += buff[d] * buff[d];
Q = (1.0 / (1.0 + Q)) / sum_Q;
C += val_P[i] * System.Math.Log((val_P[i] + FLT_MIN) / (Q + FLT_MIN));
}
}
return C;
}
// Compute input similarities with a fixed perplexity
internal static void computeGaussianPerplexity(double[][] X, int N, int D, ref double[][] P, double perplexity)
{
// Compute the squared Euclidean distance matrix
double[][] DD = Jagged.Zeros(N, N);
computeSquaredEuclideanDistance(X, N, D, DD);
// Compute the Gaussian kernel row by row
for (int n = 0; n < P.Length; n++)
{
// Initialize some variables
bool found = false;
double beta = 1.0;
double min_beta = -DBL_MAX;
double max_beta = DBL_MAX;
double tol = 1e-5;
double sum_P = 0;
// Iterate until we found a good perplexity
int iter = 0;
while (!found && iter < 200)
{
// Compute Gaussian kernel row
for (int m = 0; m < DD[n].Length; m++)
P[n][m] = System.Math.Exp(-beta * DD[n][m]);
P[n][n] = DBL_MIN;
// Compute entropy of current row
sum_P = DBL_MIN;
for (int m = 0; m < P[n].Length; m++)
sum_P += P[n][m];
double H = 0.0;
for (int m = 0; m < DD[n].Length; m++)
H += beta * (DD[n][m] * P[n][m]);
H = (H / sum_P) + System.Math.Log(sum_P);
// Evaluate whether the entropy is within the tolerance level
double Hdiff = H - System.Math.Log(perplexity);
if (Hdiff < tol && -Hdiff < tol)
{
found = true;
}
else
{
if (Hdiff > 0)
{
min_beta = beta;
if (max_beta == DBL_MAX || max_beta == -DBL_MAX)
beta *= 2.0;
else
beta = (beta + max_beta) / 2.0;
}
else
{
max_beta = beta;
if (min_beta == -DBL_MAX || min_beta == DBL_MAX)
beta /= 2.0;
else
beta = (beta + min_beta) / 2.0;
}
}
// Update iteration counter
iter++;
}
// Row normalize P
P[n].Divide(sum_P, result: P[n]);
}
}
// Compute input similarities with a fixed perplexity using ball trees (this function allocates memory another function should free)
internal static void computeGaussianPerplexity(double[][] X, int N, int D, ref int[] _row_P, ref int[] _col_P, ref double[] _val_P, double perplexity, int K)
{
if (perplexity > K)
throw new Exception(String.Format("Perplexity should be lower than K ({0}).", K));
// Allocate the memory we need
_row_P = new int[N + 1];
_col_P = new int[N * K];
_val_P = new double[N * K];
int[] row_P = _row_P;
int[] col_P = _col_P;
double[] val_P = _val_P;
double[] cur_P = new double[N - 1];
row_P[0] = 0;
for (int n = 0; n < N; n++)
row_P[n + 1] = row_P[n] + (int)K;
// Build ball tree on data set
var idx = Vector.Range(0, N);
var tree = VPTree.FromData(X, idx, inPlace: false);
// Loop over all points to find nearest neighbors
var results = new List<NodeDistance<VPTreeNode<double[], int>>>();
Debug.Write("Building tree...");
for (int n = 0; n < X.Length; n++)
{
if (n % 10000 == 0)
Debug.Write(String.Format(" - point {0} of {1}", n, N));
// Find nearest neighbors
results.Clear();
tree.Nearest(X[n], K + 1, results);
// Initialize some variables for binary search
bool found = false;
double beta = 1.0;
double min_beta = -DBL_MIN;
double max_beta = DBL_MAX;
double tol = 1e-5;
// Iterate until we found a good perplexity
int iter = 0;
double sum_P = 0;
while (!found && iter < 200)
{
// Compute Gaussian kernel row
for (int m = 0; m < K; m++)
{
double d = results[m + 1].Distance;
cur_P[m] = System.Math.Exp(-beta * d * d);
}
// Compute entropy of current row
sum_P = DBL_MIN;
for (int m = 0; m < K; m++)
sum_P += cur_P[m];
double H = 0.0;
for (int m = 0; m < K; m++)
{
double d = results[m + 1].Distance;
H += beta * (d * d * cur_P[m]);
}
H = (H / sum_P) + System.Math.Log(sum_P);
// Evaluate whether the entropy is within the tolerance level
double Hdiff = H - System.Math.Log(perplexity);
if (Hdiff < tol && -Hdiff < tol)
{
found = true;
}
else
{
if (Hdiff > 0)
{
min_beta = beta;
if (max_beta == DBL_MAX || max_beta == -DBL_MAX)
beta *= 2.0;
else
beta = (beta + max_beta) / 2.0;
}
else
{
max_beta = beta;
if (min_beta == -DBL_MAX || min_beta == DBL_MAX)
beta /= 2.0;
else
beta = (beta + min_beta) / 2.0;
}
}
// Update iteration counter
iter++;
}
// Row-normalize current row of P and store in matrix
cur_P.Divide(sum_P, result: cur_P);
for (int m = 0; m < K; m++)
{
int j = results[m + 1].Node.Value; // holds the index
col_P[row_P[n] + m] = j;
val_P[row_P[n] + m] = cur_P[m];
}
}
}
// Symmetrizes a sparse matrix
internal static void symmetrizeMatrix(ref int[] _row_P, ref int[] _col_P, ref double[] _val_P, int N)
{
// Get sparse matrix
int[] row_P = _row_P;
int[] col_P = _col_P;
double[] val_P = _val_P;
// Count number of elements and row counts of symmetric matrix
int[] row_counts = new int[N];
for (int n = 0; n < row_P.Length - 1; n++)
{
for (int i = row_P[n]; i < row_P[n + 1]; i++)
{
// Check whether element (col_P[i], n) is present
bool present = false;
for (int m = row_P[col_P[i]]; m < row_P[col_P[i] + 1]; m++)
{
if (col_P[m] == n)
present = true;
}
if (present)
{
row_counts[n]++;
}
else
{
row_counts[n]++;
row_counts[col_P[i]]++;
}
}
}
int no_elem = 0;
for (int n = 0; n < N; n++)
no_elem += row_counts[n];
// Allocate memory for symmetrized matrix
int[] sym_row_P = new int[N + 1];
int[] sym_col_P = new int[no_elem];
double[] sym_val_P = new double[no_elem];
// Construct new row indices for symmetric matrix
sym_row_P[0] = 0;
for (int n = 0; n < N; n++)
sym_row_P[n + 1] = sym_row_P[n] + (int)row_counts[n];
// Fill the result matrix
int[] offset = new int[N];
for (int n = 0; n < N; n++)
{
for (int i = row_P[n]; i < row_P[n + 1]; i++)
{
// considering element(n, col_P[i])
// Check whether element (col_P[i], n) is present
bool present = false;
for (int m = row_P[col_P[i]]; m < row_P[col_P[i] + 1]; m++)
{
if (col_P[m] == n)
{
present = true;
if (n <= col_P[i])
{
// make sure we do not add elements twice
sym_col_P[sym_row_P[n] + offset[n]] = col_P[i];
sym_col_P[sym_row_P[col_P[i]] + offset[col_P[i]]] = n;
sym_val_P[sym_row_P[n] + offset[n]] = val_P[i] + val_P[m];
sym_val_P[sym_row_P[col_P[i]] + offset[col_P[i]]] = val_P[i] + val_P[m];
}
}
}
// If (col_P[i], n) is not present, there is no addition involved
if (!present)
{
sym_col_P[sym_row_P[n] + offset[n]] = col_P[i];
sym_col_P[sym_row_P[col_P[i]] + offset[col_P[i]]] = n;
sym_val_P[sym_row_P[n] + offset[n]] = val_P[i];
sym_val_P[sym_row_P[col_P[i]] + offset[col_P[i]]] = val_P[i];
}
// Update offsets
if (!present || (present && n <= col_P[i]))
{
offset[n]++;
if (col_P[i] != n)
offset[col_P[i]]++;
}
}
}
// Divide the result by two
for (int i = 0; i < no_elem; i++)
sym_val_P[i] /= 2.0;
// Return symmetrized matrices
_row_P = sym_row_P;
_col_P = sym_col_P;
_val_P = sym_val_P;
}
// Compute squared Euclidean distance matrix (using BLAS)
internal static void computeSquaredEuclideanDistance(double[][] X, int N, int D, double[][] DD)
{
double[] dataSums = new double[N];
for (int n = 0; n < X.Length; n++)
for (int d = 0; d < X[n].Length; d++)
dataSums[n] += (X[n][d] * X[n][d]);
for (int n = 0; n < DD.Length; n++)
for (int m = 0; m < DD[n].Length; m++)
DD[n][m] = dataSums[n] + dataSums[m];
for (int n = 0; n < X.Length; n++)
{
DD[n][n] = 0.0;
for (int m = n + 1; m < X.Length; m++)
{
DD[n][m] = 0.0;
for (int d = 0; d < D; d++)
DD[n][m] += (X[n][d] - X[m][d]) * (X[n][d] - X[m][d]);
DD[m][n] = DD[n][m];
}
}
}
// Generates a Gaussian random number
private static double randn()
{
return NormalDistribution.Standard.Generate();
}
// Function that loads data from a t-SNE file
// Note: this function does a malloc that should be freed elsewhere
private static bool load_data(ref double[][] data, ref int n, ref int d, ref int no_dims, ref double theta, ref double perplexity, ref int rand_seed)
{
// Open file, read first 2 integers, allocate memory, and read the data
using (BinaryReader h = new BinaryReader(new FileStream("data.dat", FileMode.Open, FileAccess.Read)))
{
n = h.ReadInt32(); // number of datapoints
d = h.ReadInt32(); // original dimensionality
theta = h.ReadDouble(); // gradient accuracy
perplexity = h.ReadDouble(); // perplexity
no_dims = h.ReadInt32(); // output dimensionality
data = h.ReadJagged<double>(n, d);
if (h.BaseStream.Position != h.BaseStream.Length)
rand_seed = h.ReadInt32();// random seed
}
Debug.WriteLine(String.Format("Read the {0} x {1} data matrix successfully!", n, d));
return true;
}
// Function that saves map to a t-SNE file
private static void save_data(double[][] data, int[] landmarks, double[] costs, int n, int d)
{
// Open file, write first 2 integers and then the data
using (BinaryWriter h = new BinaryWriter(new FileStream("result.dat", FileMode.Create, FileAccess.Write)))
{
h.Write(n);
h.Write(d);
h.Write(data);
h.Write(landmarks);
h.Write(costs);
Debug.WriteLine(String.Format("Wrote the {0} x {1} data matrix successfully!", n, d));
}
}
}
#endif
}
Reference in New Issue View Git Blame Copy Permalink