rmontanana/Platform
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Complete excel in b_list

Browse Source
This commit is contained in:
Ricardo Montañana Gómez
2024-02-29 16:14:01 +01:00
parent b68d520726
commit 23c3bed667
4 changed files with 52 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

41
src/list/DatasetsExcel.cc
View File

@@ -4,7 +4,7 @@
namespace platform {
DatasetsExcel::DatasetsExcel()
DatasetsExcel::DatasetsExcel(json& data) : data(data), ExcelFile()
{
file_name = "datasets.xlsx";
workbook = workbook_new(getFileName().c_str());
@@ -17,34 +17,37 @@ namespace platform {
}
void DatasetsExcel::report()
{
int datasetNameSize = 25; // Min size of the column
int balanceSize = 75; // Min size of the column
worksheet = workbook_add_worksheet(workbook, "Datasets");
formatColumns();
worksheet_merge_range(worksheet, 0, 0, 0, 4, "Datasets", styles["headerFirst"]);
worksheet_merge_range(worksheet, 0, 0, 0, 5, "Datasets", styles["headerFirst"]);
formatColumns(datasetNameSize, balanceSize);
// Body header
row = 3;
int col = 1;
row = 2;
int col = 0;
int i = 0;
// Get Datasets
// auto data = platform::Datasets(false, platform::Paths::datasets());
// auto datasets = data.getNames();
auto datasets = std::vector<std::string>{ "iris", "wine", "digits", "breast_cancer" };
int maxDatasetName = (*std::max_element(datasets.begin(), datasets.end(), [](const std::string& a, const std::string& b) { return a.size() < b.size(); })).size();
datasetNameSize = std::max(datasetNameSize, maxDatasetName);
writeString(row, 0, "Nº", "bodyHeader");
writeString(row, 1, "Dataset", "bodyHeader");
for (auto const& name : datasets) {
for (const auto& name : { "Nº", "Dataset", "Samples", "Features", "Classes", "Balance" }) {
writeString(row, col++, name, "bodyHeader");
}
for (auto& [key, value] : data.items()) {
row++;
if (key.size() > datasetNameSize) {
datasetNameSize = key.size();
}
writeInt(row, 0, i++, "ints");
writeString(row, 1, name.c_str(), "text");
writeString(row, 1, key.c_str(), "text");
writeInt(row, 2, value["samples"], "ints");
writeInt(row, 3, value["features"], "ints");
writeInt(row, 4, value["classes"], "ints");
writeString(row, 5, value["balance"].get<std::string>().c_str(), "text");
}
row++;
formatColumns();
formatColumns(datasetNameSize, balanceSize);
}
void DatasetsExcel::formatColumns()
void DatasetsExcel::formatColumns(int dataset, int balance)
{
worksheet_freeze_panes(worksheet, 4, 2);
std::vector<int> columns_sizes = { 5, datasetNameSize };
std::vector<int> columns_sizes = { 5, dataset, 10, 10, 10, balance };
for (int i = 0; i < columns_sizes.size(); ++i) {
worksheet_set_column(worksheet, i, i, columns_sizes.at(i), NULL);
}

7
src/list/DatasetsExcel.h
View File

@@ -11,12 +11,13 @@ namespace platform {
class DatasetsExcel : public ExcelFile {
public:
DatasetsExcel();
explicit DatasetsExcel(json& data);
~DatasetsExcel();
void report();
private:
void formatColumns();
int datasetNameSize = 25; // Min size of the column
void formatColumns(int dataset, int balance);
json data;
};
}
#endif //DATASETS_EXCEL_H

31
src/list/b_list.cc
View File

@@ -1,6 +1,7 @@
#include <iostream>
#include <locale>
#include <argparse/argparse.hpp>
#include <nlohmann/json.hpp>
#include "Paths.h"
#include "Colors.h"
#include "Datasets.h"
@@ -15,7 +16,7 @@ struct separated : numpunct<char> {
std::string do_grouping() const { return "\03"; }
};
void outputBalance(const std::string& balance)
std::string outputBalance(const std::string& balance)
{
auto temp = std::string(balance);
while (temp.size() > BALANCE_LENGTH - 1) {
@@ -24,12 +25,12 @@ void outputBalance(const std::string& balance)
std::cout << setw(52) << " ";
temp = temp.substr(BALANCE_LENGTH);
}
std::cout << temp << std::endl;
return temp;
}
int main(int argc, char** argv)
{
auto data = platform::Datasets(false, platform::Paths::datasets());
auto datasets = platform::Datasets(false, platform::Paths::datasets());
argparse::ArgumentParser program("b_list", { project_version.begin(), project_version.end() });
program.add_argument("--excel")
.help("Output in Excel format")
@@ -44,26 +45,34 @@ int main(int argc, char** argv)
std::string balanceBars = std::string(BALANCE_LENGTH, '=');
std::cout << "=== ============================== ====== ===== === " << balanceBars << std::endl;
int num = 0;
for (const auto& dataset : data.getNames()) {
json data;
for (const auto& dataset : datasets.getNames()) {
auto color = num % 2 ? Colors::CYAN() : Colors::BLUE();
std::cout << color << setw(3) << right << num++ << " ";
std::cout << setw(30) << left << dataset << " ";
data.loadDataset(dataset);
auto nSamples = data.getNSamples(dataset);
datasets.loadDataset(dataset);
auto nSamples = datasets.getNSamples(dataset);
std::cout << setw(6) << right << nSamples << " ";
std::cout << setw(5) << right << data.getFeatures(dataset).size() << " ";
std::cout << setw(3) << right << data.getNClasses(dataset) << " ";
std::cout << setw(5) << right << datasets.getFeatures(dataset).size() << " ";
std::cout << setw(3) << right << datasets.getNClasses(dataset) << " ";
std::stringstream oss;
std::string sep = "";
for (auto number : data.getClassesCounts(dataset)) {
for (auto number : datasets.getClassesCounts(dataset)) {
oss << sep << std::setprecision(2) << fixed << (float)number / nSamples * 100.0 << "% (" << number << ")";
sep = " / ";
}
outputBalance(oss.str());
auto balance = outputBalance(oss.str());
std::cout << balance << std::endl;
// Store data for Excel report
data[dataset] = json::object();
data[dataset]["samples"] = nSamples;
data[dataset]["features"] = datasets.getFeatures(dataset).size();
data[dataset]["classes"] = datasets.getNClasses(dataset);
data[dataset]["balance"] = oss.str();
}
std::cout << Colors::RESET() << std::endl;
if (excel) {
auto report = platform::DatasetsExcel();
auto report = platform::DatasetsExcel(data);
report.report();
std::cout << "Output saved in " << report.getFileName() << std::endl;
}

6
src/reports/ExcelFile.cc
View File

@@ -87,11 +87,13 @@ namespace platform {
if (name == "textCentered") {
format_set_align(style, LXW_ALIGN_CENTER);
format_set_font_size(style, normalSize);
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_border(style, LXW_BORDER_THIN);
} else if (name == "text") {
format_set_font_size(style, normalSize);
format_set_border(style, LXW_BORDER_THIN);
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_text_wrap(style);
} else if (name == "bodyHeader") {
format_set_bold(style);
format_set_font_size(style, normalSize);
@@ -101,18 +103,22 @@ namespace platform {
format_set_bg_color(style, lxw_color_t(colorTitle));
} else if (name == "result") {
format_set_font_size(style, normalSize);
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_border(style, LXW_BORDER_THIN);
format_set_num_format(style, "0.0000000");
} else if (name == "time") {
format_set_font_size(style, normalSize);
format_set_border(style, LXW_BORDER_THIN);
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_num_format(style, "#,##0.000000");
} else if (name == "ints") {
format_set_font_size(style, normalSize);
format_set_num_format(style, "###,##0");
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_border(style, LXW_BORDER_THIN);
} else if (name == "floats") {
format_set_border(style, LXW_BORDER_THIN);
format_set_align(style, LXW_ALIGN_VERTICAL_CENTER);
format_set_font_size(style, normalSize);
format_set_num_format(style, "#,##0.00");
}