2023-01-23 21:13:26 +01:00
//
// LinReg.cpp
//
// Created by Marc Melikyan on 10/2/20.
//
2023-01-24 18:12:23 +01:00
# include "lin_reg.h"
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
# include "../cost/cost.h"
2023-01-24 18:12:23 +01:00
# include "../regularization/reg.h"
2023-01-24 19:00:54 +01:00
# include "../stat/stat.h"
2023-01-24 18:12:23 +01:00
# include "../utilities/utilities.h"
2023-01-23 21:13:26 +01:00
# include <cmath>
2023-01-24 19:00:54 +01:00
# include <iostream>
2023-01-23 21:13:26 +01:00
# include <random>
2023-02-11 11:09:29 +01:00
/*
Ref < MLPPMatrix > MLPPLinReg : : get_input_set ( ) {
return _input_set ;
}
void MLPPLinReg : : set_input_set ( const Ref < MLPPMatrix > & val ) {
_input_set = val ;
_initialized = false ;
}
Ref < MLPPVector > MLPPLinReg : : get_output_set ( ) {
return _output_set ;
}
void MLPPLinReg : : set_output_set ( const Ref < MLPPVector > & val ) {
_output_set = val ;
_initialized = false ;
}
MLPPReg : : RegularizationType MLPPLinReg : : get_reg ( ) {
return _reg ;
}
void MLPPLinReg : : set_reg ( const MLPPReg : : RegularizationType val ) {
_reg = val ;
_initialized = false ;
}
real_t MLPPLinReg : : get_lambda ( ) {
return _lambda ;
}
void MLPPLinReg : : set_lambda ( const real_t val ) {
_lambda = val ;
_initialized = false ;
}
2023-02-10 21:55:21 +01:00
2023-02-11 11:09:29 +01:00
real_t MLPPLinReg : : get_alpha ( ) {
return _alpha ;
}
void MLPPLinReg : : set_alpha ( const real_t val ) {
_alpha = val ;
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
_initialized = false ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
*/
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > MLPPLinReg : : model_set_test ( const Ref < MLPPMatrix > & X ) {
ERR_FAIL_COND_V ( ! _initialized , Ref < MLPPVector > ( ) ) ;
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
return evaluatem ( X ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-15 00:30:02 +01:00
real_t MLPPLinReg : : model_test ( const Ref < MLPPVector > & x ) {
2023-02-11 11:09:29 +01:00
ERR_FAIL_COND_V ( ! _initialized , 0 ) ;
return evaluatev ( x ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : newton_raphson ( real_t learning_rate , int max_epoch , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 00:54:50 +01:00
MLPPReg regularization ;
2023-02-11 11:09:29 +01:00
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-24 19:00:54 +01:00
int epoch = 1 ;
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
while ( true ) {
2023-02-11 11:09:29 +01:00
cost_prev = cost ( _y_hat , _output_set ) ;
2023-01-24 19:00:54 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = _y_hat - > subn ( _output_set ) ;
2023-01-24 19:00:54 +01:00
// Calculating the weight gradients (2nd derivative)
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > first_derivative = _input_set - > transposen ( ) - > mult_vec ( error ) ;
Ref < MLPPMatrix > second_derivative = _input_set - > transposen ( ) - > multn ( _input_set ) ;
_weights - > sub ( second_derivative - > inverse ( ) - > transposen ( ) - > mult_vec ( first_derivative ) - > scalar_multiplyn ( learning_rate / _n ) ) ;
2023-02-15 00:30:02 +01:00
_weights = regularization . reg_weightsv ( _weights , _lambda , _alpha , _reg ) ;
2023-01-24 19:00:54 +01:00
// Calculating the bias gradients (2nd derivative)
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / _n ; // We keep this the same. The 2nd derivative is just [1].
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( _y_hat , _output_set ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : gradient_descent ( real_t learning_rate , int max_epoch , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 00:54:50 +01:00
MLPPReg regularization ;
2023-02-11 11:09:29 +01:00
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-24 19:00:54 +01:00
int epoch = 1 ;
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
while ( true ) {
2023-02-11 11:09:29 +01:00
cost_prev = cost ( _y_hat , _output_set ) ;
2023-01-24 19:00:54 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = _y_hat - > subn ( _output_set ) ;
2023-01-24 19:00:54 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
_weights - > sub ( _input_set - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( learning_rate / _n ) ) ;
2023-02-15 00:30:02 +01:00
_weights = regularization . reg_weightsv ( _weights , _lambda , _alpha , _reg ) ;
2023-01-24 19:00:54 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / _n ;
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( _y_hat , _output_set ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : sgd ( real_t learning_rate , int max_epoch , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 00:54:50 +01:00
MLPPReg regularization ;
2023-02-11 11:09:29 +01:00
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-24 19:00:54 +01:00
int epoch = 1 ;
2023-02-11 11:09:29 +01:00
std : : random_device rd ;
std : : default_random_engine generator ( rd ( ) ) ;
std : : uniform_int_distribution < int > distribution ( 0 , int ( _n - 1 ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > input_set_row_tmp ;
input_set_row_tmp . instance ( ) ;
input_set_row_tmp - > resize ( _input_set - > size ( ) . x ) ;
Ref < MLPPVector > output_set_row_tmp ;
output_set_row_tmp . instance ( ) ;
output_set_row_tmp - > resize ( 1 ) ;
Ref < MLPPVector > y_hat_tmp ;
y_hat_tmp . instance ( ) ;
y_hat_tmp - > resize ( 1 ) ;
2023-01-24 19:00:54 +01:00
while ( true ) {
2023-02-15 00:30:02 +01:00
int output_index = distribution ( generator ) ;
2023-01-24 19:00:54 +01:00
2023-04-29 15:07:30 +02:00
_input_set - > row_get_into_mlpp_vector ( output_index , input_set_row_tmp ) ;
2023-04-29 13:44:18 +02:00
real_t output_element_set = _output_set - > element_get ( output_index ) ;
output_set_row_tmp - > element_set ( 0 , output_element_set ) ;
2023-01-24 19:00:54 +01:00
2023-02-15 00:30:02 +01:00
real_t y_hat = evaluatev ( input_set_row_tmp ) ;
2023-04-29 13:44:18 +02:00
y_hat_tmp - > element_set ( 0 , output_element_set ) ;
2023-02-15 00:30:02 +01:00
cost_prev = cost ( y_hat_tmp , output_set_row_tmp ) ;
2023-04-29 13:44:18 +02:00
real_t error = y_hat - output_element_set ;
2023-01-24 19:00:54 +01:00
// Weight updation
2023-04-30 12:51:46 +02:00
_weights - > sub ( input_set_row_tmp - > scalar_multiplyn ( learning_rate * error ) ) ;
2023-02-15 00:30:02 +01:00
_weights = regularization . reg_weightsv ( _weights , _lambda , _alpha , _reg ) ;
2023-01-24 19:00:54 +01:00
// Bias updation
2023-02-11 11:09:29 +01:00
_bias - = learning_rate * error ;
2023-01-24 19:00:54 +01:00
2023-02-15 00:30:02 +01:00
y_hat = evaluatev ( input_set_row_tmp ) ;
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat_tmp , output_set_row_tmp ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
epoch + + ;
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : mbgd ( real_t learning_rate , int max_epoch , int mini_batch_size , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 00:54:50 +01:00
MLPPReg regularization ;
2023-02-11 11:09:29 +01:00
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-24 19:00:54 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-24 19:00:54 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-24 19:00:54 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-24 19:00:54 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
_weights - > sub ( current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( learning_rate / current_output_mini_batch - > size ( ) ) ) ;
2023-02-15 00:30:02 +01:00
_weights = regularization . reg_weightsv ( _weights , _lambda , _alpha , _reg ) ;
2023-01-24 19:00:54 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ;
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-24 19:00:54 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-24 19:00:54 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-24 19:00:54 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : momentum ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t gamma , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Momentum.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
v = v - > scalar_multiplyn ( gamma ) - > addn ( weight_grad - > scalar_multiplyn ( learning_rate ) ) ;
_weights - > sub ( v ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : nag ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t gamma , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Momentum.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
_weights - > sub ( v - > scalar_multiplyn ( gamma ) ) ; // "Aposterori" calculation
2023-01-25 23:43:21 +01:00
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
v = v - > scalar_multiplyn ( gamma ) - > addn ( weight_grad - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
_weights - > sub ( v ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : adagrad ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t e , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Adagrad.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
v = weight_grad - > hadamard_productn ( weight_grad ) ;
_weights - > sub ( weight_grad - > division_element_wisen ( v - > scalar_addn ( e ) - > sqrtn ( ) ) - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : adadelta ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t b1 , real_t e , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
// Adagrad upgrade. Momentum is applied.
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Adagrad.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
v = v - > scalar_multiplyn ( b1 ) - > addn ( weight_grad - > hadamard_productn ( weight_grad ) - > scalar_multiplyn ( 1 - b1 ) ) ;
_weights - > sub ( weight_grad - > division_element_wisen ( v - > scalar_addn ( e ) - > sqrtn ( ) ) - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : adam ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t b1 , real_t b2 , real_t e , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Adam.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
m = m - > scalar_multiplyn ( b1 ) - > addn ( weight_grad - > scalar_multiplyn ( 1 - b1 ) ) ;
v = v - > scalar_multiplyn ( b2 ) - > addn ( weight_grad - > exponentiaten ( 2 ) - > scalar_multiplyn ( 1 - b2 ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m_hat = m - > scalar_multiplyn ( 1 / ( 1 - Math : : pow ( b1 , epoch ) ) ) ;
Ref < MLPPVector > v_hat = v - > scalar_multiplyn ( 1 / ( 1 - Math : : pow ( b2 , epoch ) ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
_weights - > sub ( m_hat - > division_element_wisen ( v_hat - > sqrtn ( ) - > scalar_addn ( e ) ) - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : adamax ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t b1 , real_t b2 , real_t e , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
Ref < MLPPVector > u = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
2023-01-25 23:43:21 +01:00
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
m = m - > scalar_multiplyn ( b1 ) - > addn ( weight_grad - > scalar_multiplyn ( 1 - b1 ) ) ;
u = u - > scalar_multiplyn ( b2 ) - > maxn ( weight_grad - > absn ( ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m_hat = m - > scalar_multiplyn ( 1 / ( 1 - Math : : pow ( b1 , epoch ) ) ) ;
_weights - > sub ( m_hat - > division_element_wisen ( u ) - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : nadam ( real_t learning_rate , int max_epoch , int mini_batch_size , real_t b1 , real_t b2 , real_t e , bool ui ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPReg regularization ;
2023-01-27 13:01:16 +01:00
real_t cost_prev = 0 ;
2023-01-25 23:43:21 +01:00
int epoch = 1 ;
// Creating the mini-batches
2023-02-11 11:09:29 +01:00
int n_mini_batch = _n / mini_batch_size ;
2023-02-15 00:30:02 +01:00
MLPPUtilities : : CreateMiniBatchMVBatch batches = MLPPUtilities : : create_mini_batchesmv ( _input_set , _output_set , n_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Initializing necessary components for Adam.
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
Ref < MLPPVector > v = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
Ref < MLPPVector > m_final = MLPPVector : : create_vec_zero ( _weights - > size ( ) ) ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
while ( true ) {
for ( int i = 0 ; i < n_mini_batch ; i + + ) {
2023-02-15 00:30:02 +01:00
Ref < MLPPMatrix > current_input_mini_batch = batches . input_sets [ i ] ;
Ref < MLPPVector > current_output_mini_batch = batches . output_sets [ i ] ;
Ref < MLPPVector > y_hat = evaluatem ( current_input_mini_batch ) ;
cost_prev = cost ( y_hat , current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > error = y_hat - > subn ( current_output_mini_batch ) ;
2023-01-25 23:43:21 +01:00
// Calculating the weight gradients
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > gradient = current_input_mini_batch - > transposen ( ) - > mult_vec ( error ) - > scalar_multiplyn ( 1 / current_output_mini_batch - > size ( ) ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > reg_deriv_term = regularization . reg_deriv_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > weight_grad = gradient - > addn ( reg_deriv_term ) ; // Weight_grad_final
m = m - > scalar_multiplyn ( b1 ) - > addn ( weight_grad - > scalar_multiplyn ( 1 - b1 ) ) ;
v = v - > scalar_multiplyn ( b2 ) - > addn ( weight_grad - > exponentiaten ( 2 ) - > scalar_multiplyn ( 1 - b2 ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
m_final = m - > scalar_multiplyn ( b1 ) - > addn ( weight_grad - > scalar_multiplyn ( ( 1 - b1 ) / ( 1 - Math : : pow ( b1 , epoch ) ) ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPVector > m_hat = m - > scalar_multiplyn ( 1 / ( 1 - Math : : pow ( b1 , epoch ) ) ) ;
Ref < MLPPVector > v_hat = v - > scalar_multiplyn ( 1 / ( 1 - Math : : pow ( b2 , epoch ) ) ) ;
2023-01-25 23:43:21 +01:00
2023-04-30 12:51:46 +02:00
_weights - > sub ( m_final - > division_element_wisen ( v_hat - > sqrtn ( ) - > scalar_addn ( e ) ) - > scalar_multiplyn ( learning_rate ) ) ;
2023-01-25 23:43:21 +01:00
// Calculating the bias gradients
2023-04-30 12:51:46 +02:00
_bias - = learning_rate * error - > sum_elements ( ) / current_output_mini_batch - > size ( ) ; // As normal
2023-02-15 00:30:02 +01:00
y_hat = evaluatem ( current_input_mini_batch ) ;
2023-01-25 23:43:21 +01:00
2023-02-11 11:09:29 +01:00
if ( ui ) {
2023-02-15 00:30:02 +01:00
MLPPUtilities : : cost_info ( epoch , cost_prev , cost ( y_hat , current_output_mini_batch ) ) ;
MLPPUtilities : : print_ui_vb ( _weights , _bias ) ;
2023-01-25 23:43:21 +01:00
}
}
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
epoch + + ;
2023-02-11 11:09:29 +01:00
2023-01-25 23:43:21 +01:00
if ( epoch > max_epoch ) {
break ;
}
}
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : normal_equation ( ) {
ERR_FAIL_COND ( ! _initialized ) ;
2023-01-25 23:43:21 +01:00
MLPPStat stat ;
2023-01-24 19:00:54 +01:00
2023-04-30 12:51:46 +02:00
Ref < MLPPMatrix > input_set_t = _input_set - > transposen ( ) ;
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > input_set_t_row_tmp ;
input_set_t_row_tmp . instance ( ) ;
input_set_t_row_tmp - > resize ( input_set_t - > size ( ) . x ) ;
Ref < MLPPVector > x_means ;
x_means . instance ( ) ;
x_means - > resize ( input_set_t - > size ( ) . y ) ;
for ( int i = 0 ; i < input_set_t - > size ( ) . y ; i + + ) {
2023-04-29 15:07:30 +02:00
input_set_t - > row_get_into_mlpp_vector ( i , input_set_t_row_tmp ) ;
2023-02-15 00:30:02 +01:00
2023-04-29 13:44:18 +02:00
x_means - > element_set ( i , stat . meanv ( input_set_t_row_tmp ) ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > temp ;
//temp.resize(_k);
2023-04-30 12:51:46 +02:00
temp = _input_set - > transposen ( ) - > multn ( _input_set ) - > inverse ( ) - > mult_vec ( _input_set - > transposen ( ) - > mult_vec ( _output_set ) ) ;
2023-01-24 19:00:54 +01:00
2023-04-29 13:44:18 +02:00
ERR_FAIL_COND_MSG ( Math : : is_nan ( temp - > element_get ( 0 ) ) , " ERR: Resulting matrix was noninvertible/degenerate, and so the normal equation could not be performed. Try utilizing gradient descent. " ) ;
2023-01-24 19:00:54 +01:00
2023-02-15 00:30:02 +01:00
if ( _reg = = MLPPReg : : REGULARIZATION_TYPE_RIDGE ) {
2023-04-30 12:51:46 +02:00
_weights = _input_set - > transposen ( ) - > multn ( _input_set ) - > addn ( MLPPMatrix : : create_identity_mat ( _k ) - > scalar_multiplyn ( _lambda ) ) - > inverse ( ) - > mult_vec ( _input_set - > transposen ( ) - > mult_vec ( _output_set ) ) ;
2023-02-11 11:09:29 +01:00
} else {
2023-04-30 12:51:46 +02:00
_weights = _input_set - > transposen ( ) - > multn ( _input_set ) - > inverse ( ) - > mult_vec ( _input_set - > transposen ( ) - > mult_vec ( _output_set ) ) ;
2023-01-25 23:43:21 +01:00
}
2023-02-11 11:09:29 +01:00
2023-04-30 12:51:46 +02:00
_bias = stat . meanv ( _output_set ) - _weights - > dot ( x_means ) ;
2023-02-11 11:09:29 +01:00
forward_pass ( ) ;
2023-01-24 19:00:54 +01:00
}
2023-01-27 13:01:16 +01:00
real_t MLPPLinReg : : score ( ) {
2023-02-11 11:09:29 +01:00
ERR_FAIL_COND_V ( ! _initialized , 0 ) ;
2023-01-25 23:43:21 +01:00
MLPPUtilities util ;
2023-02-11 11:09:29 +01:00
2023-02-15 00:30:02 +01:00
return util . performance_vec ( _y_hat , _output_set ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-15 00:30:02 +01:00
void MLPPLinReg : : save ( const String & file_name ) {
2023-02-11 11:09:29 +01:00
ERR_FAIL_COND ( ! _initialized ) ;
2023-02-15 00:30:02 +01:00
//MLPPUtilities util;
2023-02-11 11:09:29 +01:00
2023-02-15 00:30:02 +01:00
//util.saveParameters(fileName, _weights, _bias);
2023-01-24 19:00:54 +01:00
}
2023-02-11 11:09:29 +01:00
bool MLPPLinReg : : is_initialized ( ) {
return _initialized ;
}
void MLPPLinReg : : initialize ( ) {
if ( _initialized ) {
return ;
}
//ERR_FAIL_COND(!_input_set.is_valid() || !_output_set.is_valid());
_initialized = true ;
}
2023-02-15 00:30:02 +01:00
MLPPLinReg : : MLPPLinReg ( const Ref < MLPPMatrix > & p_input_set , const Ref < MLPPVector > & p_output_set , MLPPReg : : RegularizationType p_reg , real_t p_lambda , real_t p_alpha ) {
2023-02-11 11:09:29 +01:00
_input_set = p_input_set ;
_output_set = p_output_set ;
2023-02-15 00:30:02 +01:00
_n = p_input_set - > size ( ) . y ;
_k = p_input_set - > size ( ) . x ;
2023-02-11 11:09:29 +01:00
_reg = p_reg ;
_lambda = p_lambda ;
_alpha = p_alpha ;
2023-02-15 00:30:02 +01:00
_y_hat . instance ( ) ;
_y_hat - > resize ( _n ) ;
_weights . instance ( ) ;
_weights - > resize ( _k ) ;
MLPPUtilities utils ;
2023-02-11 11:09:29 +01:00
2023-02-15 00:30:02 +01:00
utils . weight_initializationv ( _weights ) ;
_bias = utils . bias_initializationr ( ) ;
2023-02-11 11:09:29 +01:00
_initialized = true ;
}
MLPPLinReg : : MLPPLinReg ( ) {
_initialized = false ;
}
MLPPLinReg : : ~ MLPPLinReg ( ) {
}
2023-02-15 00:30:02 +01:00
real_t MLPPLinReg : : cost ( const Ref < MLPPVector > & y_hat , const Ref < MLPPVector > & y ) {
2023-01-25 00:54:50 +01:00
MLPPReg regularization ;
2023-02-11 11:09:29 +01:00
MLPPCost mlpp_cost ;
2023-02-15 00:30:02 +01:00
return mlpp_cost . msev ( y_hat , y ) + regularization . reg_termv ( _weights , _lambda , _alpha , _reg ) ;
2023-01-24 19:00:54 +01:00
}
2023-02-15 00:30:02 +01:00
real_t MLPPLinReg : : evaluatev ( const Ref < MLPPVector > & x ) {
2023-04-30 12:51:46 +02:00
return _weights - > dot ( x ) + _bias ;
2023-01-24 19:00:54 +01:00
}
2023-02-15 00:30:02 +01:00
Ref < MLPPVector > MLPPLinReg : : evaluatem ( const Ref < MLPPMatrix > & X ) {
2023-04-30 12:51:46 +02:00
return X - > mult_vec ( _weights ) - > scalar_addn ( _bias ) ;
2023-01-24 19:00:54 +01:00
}
// wTx + b
2023-02-11 11:09:29 +01:00
void MLPPLinReg : : forward_pass ( ) {
_y_hat = evaluatem ( _input_set ) ;
}
void MLPPLinReg : : _bind_methods ( ) {
/*
ClassDB : : bind_method ( D_METHOD ( " get_input_set " ) , & MLPPLinReg : : get_input_set ) ;
ClassDB : : bind_method ( D_METHOD ( " set_input_set " , " val " ) , & MLPPLinReg : : set_input_set ) ;
ADD_PROPERTY ( PropertyInfo ( Variant : : OBJECT , " input_set " , PROPERTY_HINT_RESOURCE_TYPE , " MLPPMatrix " ) , " set_input_set " , " get_input_set " ) ;
ClassDB : : bind_method ( D_METHOD ( " get_output_set " ) , & MLPPLinReg : : get_output_set ) ;
ClassDB : : bind_method ( D_METHOD ( " set_output_set " , " val " ) , & MLPPLinReg : : set_output_set ) ;
ADD_PROPERTY ( PropertyInfo ( Variant : : OBJECT , " output_set " , PROPERTY_HINT_RESOURCE_TYPE , " MLPPVector " ) , " set_output_set " , " get_output_set " ) ;
ClassDB : : bind_method ( D_METHOD ( " get_reg " ) , & MLPPLinReg : : get_reg ) ;
ClassDB : : bind_method ( D_METHOD ( " set_reg " , " val " ) , & MLPPLinReg : : set_reg ) ;
ADD_PROPERTY ( PropertyInfo ( Variant : : INT , " reg " ) , " set_reg " , " get_reg " ) ;
ClassDB : : bind_method ( D_METHOD ( " get_lambda " ) , & MLPPLinReg : : get_lambda ) ;
ClassDB : : bind_method ( D_METHOD ( " set_lambda " , " val " ) , & MLPPLinReg : : set_lambda ) ;
ADD_PROPERTY ( PropertyInfo ( Variant : : REAL , " lambda " ) , " set_lambda " , " get_lambda " ) ;
ClassDB : : bind_method ( D_METHOD ( " get_alpha " ) , & MLPPLinReg : : get_alpha ) ;
ClassDB : : bind_method ( D_METHOD ( " set_alpha " , " val " ) , & MLPPLinReg : : set_alpha ) ;
ADD_PROPERTY ( PropertyInfo ( Variant : : REAL , " alpha " ) , " set_alpha " , " get_alpha " ) ;
ClassDB : : bind_method ( D_METHOD ( " model_test " , " x " ) , & MLPPLinReg : : model_test ) ;
ClassDB : : bind_method ( D_METHOD ( " model_set_test " , " X " ) , & MLPPLinReg : : model_set_test ) ;
ClassDB : : bind_method ( D_METHOD ( " gradient_descent " , " learning_rate " , " max_epoch " , " ui " ) , & MLPPLinReg : : gradient_descent , false ) ;
ClassDB : : bind_method ( D_METHOD ( " sgd " , " learning_rate " , " max_epoch " , " ui " ) , & MLPPLinReg : : sgd , false ) ;
ClassDB : : bind_method ( D_METHOD ( " mbgd " , " learning_rate " , " max_epoch " , " mini_batch_size " , " ui " ) , & MLPPLinReg : : mbgd , false ) ;
ClassDB : : bind_method ( D_METHOD ( " score " ) , & MLPPLinReg : : score ) ;
ClassDB : : bind_method ( D_METHOD ( " save " , " file_name " ) , & MLPPLinReg : : save ) ;
ClassDB : : bind_method ( D_METHOD ( " is_initialized " ) , & MLPPLinReg : : is_initialized ) ;
ClassDB : : bind_method ( D_METHOD ( " initialize " ) , & MLPPLinReg : : initialize ) ;
*/
2023-01-24 19:00:54 +01:00
}