Commit 3011992b authored by Josiah Grimes's avatar Josiah Grimes
Browse files

Work on enum refactoring

parent dd5b66bf

.gitignore

+0 −1
Original line number Diff line number Diff line
@@ -21,5 +21,4 @@ cv_* 
/include/

/cmake-build-*/



docker/kitware-archive.sh

src/external/pybind11

build_third_party.bash

100755 → 100644
+0 −0

File mode changed from 100755 to 100644.

View file

docker/kitware-archive.sh

0 → 100644
+103 −0
Original line number Diff line number Diff line 
#!/bin/sh



set -eu



help() {

  echo "Usage: $0 [--release <ubuntu-release>] [--rc]" > /dev/stderr

}



doing=

rc=

release=

help=

for opt in "$@"

do

  case "${doing}" in

  release)

    release="${opt}"

    doing=

    ;;

  "")

    case "${opt}" in

    --rc)

      rc=1

      ;;

    --release)

      doing=release

      ;;

    --help)

      help=1

      ;;

    esac

    ;;

  esac

done



if [ -n "${doing}" ]

then

  echo "--${doing} option given no argument." > /dev/stderr

  echo > /dev/stderr

  help

  exit 1

fi



if [ -n "${help}" ]

then

  help

  exit

fi



if [ -z "${release}" ]

then

  unset UBUNTU_CODENAME

  . /etc/os-release



  if [ -z "${UBUNTU_CODENAME+x}" ]

  then

    echo "This is not an Ubuntu system. Aborting." > /dev/stderr

    exit 1

  fi



  release="${UBUNTU_CODENAME}"

fi



case "${release}" in

xenial)

  packages="apt-transport-https"

  keyring_packages="wget"

  ;;

bionic|focal)

  packages=

  keyring_packages="gpg wget"

  ;;

*)

  echo "Only Ubuntu Xenial (16.04), Bionic (18.04), and Focal (20.04) are supported. Aborting." > /dev/stderr

  exit 1

  ;;

esac



get_keyring=

if [ ! -f /usr/share/keyrings/kitware-archive-keyring.gpg ]

then

  packages="${packages} ${keyring_packages}"

  get_keyring=1

fi



# Start the real work

set -x



apt-get update

# shellcheck disable=SC2086

apt-get install -y ${packages}



test -n "${get_keyring}" && (wget -O - https://apt.kitware.com/keys/kitware-archive-latest.asc 2>/dev/null | gpg --dearmor - > /usr/share/keyrings/kitware-archive-keyring.gpg)



echo "deb [signed-by=/usr/share/keyrings/kitware-archive-keyring.gpg] https://apt.kitware.com/ubuntu/ ${release} main" > /etc/apt/sources.list.d/kitware.list

if [ -n "${rc}" ]

then

  echo "deb [signed-by=/usr/share/keyrings/kitware-archive-keyring.gpg] https://apt.kitware.com/ubuntu/ ${release}-rc main" >> /etc/apt/sources.list.d/kitware.list

fi



apt-get update

test -n "${get_keyring}" && rm /usr/share/keyrings/kitware-archive-keyring.gpg

apt-get install -y kitware-archive-keyring

src/descriptor_identifier/model/Model.cpp

+32 −27
Original line number Diff line number Diff line
@@ -20,8 +20,8 @@ 
 */



#include "descriptor_identifier/model/Model.hpp"

#include <filesystem>

#include <utility>







Model::Model(const std::string& prop_label,

             const Unit& prop_unit,
@@ -31,7 +31,8 @@ Model::Model(const std::string& prop_label, 
             const std::vector<std::string>& sample_ids_train,

             const std::vector<std::string>& sample_ids_test,

             const std::vector<std::string>& task_names,

             const std::map<int, int>& index_map)

             const std::map<int, int>& index_map,

             const std::string loss_type_input)

    : _index_map(index_map),

      _sample_ids_train(sample_ids_train),

      _sample_ids_test(sample_ids_test),
@@ -44,7 +45,9 @@ Model::Model(const std::string& prop_label, 
      _prop_label(prop_label),

      _prop_unit(prop_unit),

      _task_eval(task_names[0]),

      _fix_intercept(loss->fix_intercept())

      _fix_intercept(loss->fix_intercept()),

      _loss_type(get_loss_type_from_string(loss_type_input))



{

    for (size_t tn = 0; tn < task_names.size(); ++tn)

    {
@@ -73,7 +76,8 @@ Model::Model(std::string prop_label, 
             std::vector<std::string> sample_ids_train,

             std::vector<std::string> sample_ids_test,

             std::vector<std::string> task_names,

             bool fix_intercept)

             bool fix_intercept,

             std::string loss_type_input)

    : _sample_ids_train(std::move(sample_ids_train)),

      _sample_ids_test(std::move(sample_ids_test)),

      _n_samp_train(feats[0]->n_samp()),
@@ -84,7 +88,8 @@ Model::Model(std::string prop_label, 
      _prop_label(std::move(prop_label)),

      _prop_unit(prop_unit),

      _task_eval(task_names[0]),

      _fix_intercept(fix_intercept)

      _fix_intercept(fix_intercept),

      _loss_type(get_loss_type_from_string(loss_type_input))

{

    for(int ii = 0; ii < _n_samp_train; ++ii)

    {
@@ -761,17 +766,10 @@ void Model::populate_model(const std::string& train_filename, const std::string& 
    int n_dim = 0;

    std::getline(train_file_stream, line);



    //if the loss_type is unknown just grab it from the error line, 3rd line down

    if (_loss_type == ""){

           _loss_type = grab_loss_type_from_line(error_line);

    }



    do

    {

        split_line = str_utils::split_string_trim(line);

        int n_tokens = static_cast<int>(split_line.size());

        n_dim = n_tokens - 3 + _fix_intercept;

        //n_dim = split_line.size() - 3 + _fix_intercept;

        n_dim = split_line.size() - 3 + _fix_intercept;

        _coefs.push_back(std::vector<double>(n_dim + (!_fix_intercept), 0.0));

        std::transform(

            split_line.begin() + 1, split_line.end() - 1, _coefs.back().data(), [](const std::string& s) {
@@ -813,18 +811,22 @@ void Model::populate_model(const std::string& train_filename, const std::string& 
        split_line = str_utils::split_string_trim(line);

        _task_names[split_line[0].substr(2)] = n_task;

        ++n_task;

//        for (std::string str : split_line){

//            std::cout << "split line piece " << str << std::endl;

//        }

//        std::string fixed_line;

//        for (char i : split_line[1])

//        {

//            if (std::isdigit(i))

//            {

//                fixed_line += i;

//            }

//        }

//        _n_samp_train += std::stoi(fixed_line);

//        task_sizes_train.push_back(std::stoi(fixed_line));

        _n_samp_train += std::stoi(split_line[1]);

        task_sizes_train.push_back(std::stoi(split_line[1]));



        std::string fixed_line;

        for (char i : split_line[1])

        {

            if (std::isdigit(i))

            {

                fixed_line += i;

            }

        }

        std::cout << "fixed line " << fixed_line << "\n";

        _n_samp_train += std::stoi(fixed_line);

        task_sizes_train.push_back(std::stoi(fixed_line));

        if (with_test)

        {

            split_line = str_utils::split_string_trim(test_line);
@@ -1141,9 +1143,9 @@ std::string Model::write_for_test_file() const{ 
    std::filesystem::path base_dir = source_path.parent_path();

    //we can reroute this to another place in repo if desired

    //but it should be killed by boost anyways after the tests are run

    std::string outpath = (base_dir / (this->_loss_type + "_model_output.txt"));

    std::string outpath = (base_dir / (get_loss_type_from_enum(this->_loss_type) + "_model_output.txt"));

    this->to_file(outpath, true);

    return (base_dir / (this->_loss_type + "_model_output.txt")).string();

    return (base_dir / (get_loss_type_from_enum(this->_loss_type) + "_model_output.txt")).string();

};



void Model::reorder_values(std::map<int, int> index_map)
@@ -1227,3 +1229,6 @@ std::string grab_loss_type_from_file(const std::string& path) { 
}

src/descriptor_identifier/model/Model.hpp

+14 −10
Original line number Diff line number Diff line
@@ -26,9 +26,9 @@ 
#include <fstream>

#include <sstream>

#include <iomanip>



#include <filesystem>

#include "loss_function/utils.hpp"



#include "utils/enum.hpp"

#ifdef PY_BINDINGS

namespace py = pybind11;

#endif
@@ -59,7 +59,7 @@ protected: 
    Unit _prop_unit; //!< The Unit for the property



    std::string _task_eval; //!< Which set of coefficients to use for evaluating the model for a new data set

    std::string _loss_type; //!< Stores the loss type of the model

    LOSS_TYPE _loss_type; //!< Stores the loss type of the model



    bool _fix_intercept; //!< If true fix intercept to 0

    // clang-format on
@@ -91,7 +91,8 @@ public: 
          const std::vector<std::string>& sample_ids_train,

          const std::vector<std::string>& sample_ids_test,

          const std::vector<std::string>& task_names,

          const std::map<int, int>& index_map);

          const std::map<int, int>& index_map,

          const std::string loss_type_input);



    // DocString: model_init_pickle

    /**
@@ -121,7 +122,8 @@ public: 
          std::vector<std::string> sample_ids_train,

          std::vector<std::string> sample_ids_test,

          std::vector<std::string> task_names,

          bool fix_intercept);

          bool fix_intercept,

          const std::string loss_type_input);



    /**

     * @brief Copy Constructor
@@ -543,7 +545,7 @@ public: 
    inline bool fix_intercept() const { return _fix_intercept; }



    inline std::string get_loss_type() const {

        return _loss_type;

        return get_loss_type_from_enum(_loss_type);

    };



    // DocString: model_coefs
@@ -602,10 +604,6 @@ public: 
                   std::string row_names="Materials");





    inline void set_loss_type(std::string new_loss_type){

        _loss_type = new_loss_type;

    }



    /**

     * @brief The property vector for all of the training samples

     */
@@ -886,6 +884,12 @@ public: 
    }

#endif

};













    /**

     * @brief Grabs the loss type from a file

     * @param line The error line within the file