shikharbhardwaj

################################################################################
With 1 random shuffle, before iteration.
################################################################################
λ hogwild/mc ∴ ./a.out -v
Thead share size : 25049701
[INFO ] Called specialization
[INFO ] Start computing loss for iteration : 1
[INFO ] Parallel SGD: iteration 1, objective 1.82124e+09.
[INFO ] Number of elements to shuffle : 100198806
[INFO ] Start shuffle for iteration : 1

shikharbhardwaj

λ mlpack_spike/hogwild ∴ g++ svm_main.cpp -O2 -std=c++11 -Wall -larmadillo -lmlpack -fopenmp
λ mlpack_spike/hogwild ∴ ./a.out
RMSE : 0.797936
Initial loss : 70176.4
Final loss : 114.332
RMSE : 0.050055
λ mlpack_spike/hogwild ∴ ./a.out
RMSE : 0.797936
Initial loss : 70176.4
Final loss : 114.369

shikharbhardwaj

#include <mlpack/core.hpp>
#include <mlpack/core/optimizers/parallel_sgd/parallel_sgd.hpp>
#include <mlpack/core/optimizers/parallel_sgd/decay_policies/constant_step.hpp>
#include <mlpack/core/optimizers/parallel_sgd/sparse_mc_function.hpp>

using namespace std;
using namespace mlpack;
using namespace mlpack::optimization;
int main()
{

shikharbhardwaj

#include <mlpack/core.hpp>
#include <mlpack/core/optimizers/parallel_sgd/parallel_sgd.hpp>
#include <mlpack/core/optimizers/parallel_sgd/decay_policies/constant_step.hpp>
#include <mlpack/core/optimizers/parallel_sgd/sparse_mc_function.hpp>

using namespace std;
using namespace mlpack::optimization;
int main()
{
  size_t numRows, numCols;

shikharbhardwaj

#include <mlpack/core.hpp>
#include <mlpack/core/optimizers/parallel_sgd/decay_policies/constant_step.hpp>
#include <mlpack/core/optimizers/parallel_sgd/parallel_sgd.hpp>
#include <mlpack/core/optimizers/parallel_sgd/sparse_mc_function.hpp>

using namespace std;
using namespace mlpack;
using namespace mlpack::optimization;
int main(int argc, char* argv[])
{

shikharbhardwaj

#include <armadillo>
#include <iostream>

int main() {
    // Load the data from the file to memory
    const size_t num_examples = 23149, num_features = 47236;
    const size_t num_locations = 1780950;

    // Lets first get the data in dense form
    arma::umat locations(2, num_locations);

shikharbhardwaj

#include <atomic>

// Maintain the decison variable as std::atomic
std::atomic<float> x[10];

void update(std::atomic<float> &f, float val){
	float old;
	do {
		old = f.load();
	}

shikharbhardwaj

#!/usr/bin/env python
import numpy as np
import matplotlib.pyplot as plt

def showKMeans(data, clusters):
    plt.clf()
    plt.scatter(data[:, 0], data[:, 1])
    plt.scatter(clusters[:, 0], clusters[:, 1])
    plt.show()

shikharbhardwaj

<?xml version="1.0" encoding="UTF-8"?>
<Site BuildName="(empty)"
	BuildStamp="20170607-1102-Experimental"
	Name="(empty)"
	Generator="ctest-3.7.2"
	CompilerName=""
	CompilerVersion=""
	OSName="Linux"
	Hostname="workstation"
	OSRelease="4.10.0-21-generic"

shikharbhardwaj

/**
 * @file naive_bayes_classifier_impl.hpp
 * @author Parikshit Ram (pram@cc.gatech.edu)
 * @author Vahab Akbarzadeh (v.akbarzadeh@gmail.com)
 * @author Shihao Jing (shihao.jing810@gmail.com)
 *
 * A Naive Bayes Classifier which parametrically estimates the distribution of
 * the features.  This classifier makes its predictions based on the assumption
 * that the features have been sampled from a set of Gaussians with diagonal
 * covariance.