Data.h

Go to the documentation of this file.

#ifndef DATA_H
#define DATA_H
 
#include "CNum/DataStructs/DataStructs.h"
 
#include <string>
#include <memory>
#include <fstream>
#include <sstream>
#include <array>

namespace CNum::Data {
  struct Bin {
    uint32_t num;
    size_t ct;
    Bin() : ct(0) {}
  };

  struct Shelf {
    size_t num_bins;
    std::unique_ptr<Bin[]> bins;
    std::unique_ptr<double[]> ranges;
 
    Shelf() : num_bins(0) {}
    Shelf(size_t nb)
      : num_bins(nb),
    bins(std::make_unique<Bin[]>(nb)),
    ranges(std::make_unique<double[]>(nb - 1)) {}
 
    Shelf &operator=(Shelf &&other) {
      if (this == &other)
    return *this;
    
      num_bins = other.num_bins;
 
      bins.reset();
      ranges.reset();
    
      bins = std::move(other.bins);
      ranges = std::move(other.ranges);
 
      return *this;
    }
  };
  
  std::array< CNum::DataStructs::Matrix<double>, 2 > get_data(std::string data_path, char seperator = ',');

  void PCA(std::string input_path, std::string output_path);

  std::shared_ptr<Shelf[]> uniform_bin(const CNum::DataStructs::Matrix<double> &data, size_t num_bins = 256);

  std::shared_ptr<Shelf[]> quantile_bin(const CNum::DataStructs::Matrix<double> &data, size_t num_bins = 256);

  CNum::DataStructs::Matrix<int> apply_quantile(const CNum::DataStructs::Matrix<double> &data, std::shared_ptr<Shelf[]> shelves);
};
 
#endif