dl::utils::Preprocessor Class Reference

#include <data_loader.hpp>

Static Public Member Functions
static MatrixD	normalize (const MatrixD &data, double min_val=0.0, double max_val=1.0)
	Normalize data to specified range.
static MatrixD	standardize (const MatrixD &data)
	Standardize data to zero mean and unit variance.
static MatrixD	one_hot_encode (const std::vector< int > &labels, size_t num_classes)
	Convert categorical labels to one-hot encoding.
static std::tuple< Dataset< double >, Dataset< double >, Dataset< double > >	train_val_test_split (const Dataset< double > &data, double train_ratio=0.7, double val_ratio=0.15)
	Split dataset into training, validation, and test sets.

Detailed Description

Definition at line 280 of file data_loader.hpp.

Member Function Documentation

normalize()

MatrixD dl::utils::Preprocessor::normalize ( const MatrixD &  data, double  min_val = 0.0, double  max_val = 1.0  )

static

Normalize data to specified range.

Parameters

data	Input data matrix
min_val	Minimum value of output range
max_val	Maximum value of output range

Returns

Normalized data matrix

Note

Uses min-max normalization: (x - min) / (max - min) * (max_val - min_val) + min_val

Examples

/home/runner/work/deep-learning-algo-impls/deep-learning-algo-impls/include/utils/data_loader.hpp.

Definition at line 178 of file data_loader.cpp.

one_hot_encode()

MatrixD dl::utils::Preprocessor::one_hot_encode ( const std::vector< int > &  labels, size_t  num_classes  )

static

Convert categorical labels to one-hot encoding.

Parameters

labels	Vector of integer class labels
num_classes	Total number of classes

Returns

Matrix with one-hot encoded labels

Note

Each row represents one sample, columns represent classes

Examples

/home/runner/work/deep-learning-algo-impls/deep-learning-algo-impls/include/utils/data_loader.hpp.

Definition at line 200 of file data_loader.cpp.

standardize()

MatrixD dl::utils::Preprocessor::standardize ( const MatrixD &  data )

static

Standardize data to zero mean and unit variance.

Parameters

data	Input data matrix

Returns

Standardized data matrix

Note

Uses z-score normalization: (x - mean) / std_dev

Examples

/home/runner/work/deep-learning-algo-impls/deep-learning-algo-impls/include/utils/data_loader.hpp.

Definition at line 189 of file data_loader.cpp.

train_val_test_split()

std::tuple< Dataset< double >, Dataset< double >, Dataset< double > > dl::utils::Preprocessor::train_val_test_split ( const Dataset< double > &  data, double  train_ratio = 0.7, double  val_ratio = 0.15  )

static

Split dataset into training, validation, and test sets.

Parameters

data	Input dataset to split
train_ratio	Fraction of data for training (default: 0.7)
val_ratio	Fraction of data for validation (default: 0.15)

Returns

Tuple of (training, validation, test) datasets

Note

Test ratio is automatically calculated as 1 - train_ratio - val_ratio

Examples

/home/runner/work/deep-learning-algo-impls/deep-learning-algo-impls/include/utils/data_loader.hpp.

Definition at line 216 of file data_loader.cpp.

---

The documentation for this class was generated from the following files:

• include/utils/data_loader.hpp
• src/utils/data_loader.cpp