tests/testthat/test-DFML_gradientBoosting.R
In jdestefani/ExtendedDFML: Extended Dynamic Factor Machine Learner
context("DFML - Test gradient boosting")
library(keras)
library(dplyr)
library(forecast)
library(parallel)
library(xgboost)
library(lightgbm)
library(MEMTS)
#Set up - Data
X <- read.csv("testdata/Sigma4.ssv",sep=" ",header = TRUE)
X <- as.matrix(X)
splitting_point <- round(2*nrow(X)/3)
X_train <- scale(X[1:splitting_point,])
components <- 3
horizon <- 5
time_window <- horizon
ss_results.df <- data.frame(DimensionalityMethod=character(),ForecastingMethod=character(),Dataset=character(),Horizon=numeric(),Columns=numeric(),Time=numeric(),MSE=numeric(),Samples=numeric(),stringsAsFactors = FALSE)
# If there is an error with the Arima package, check
#detach("package:dse", unload=TRUE)
test_that("[DFML] - PCA + gradientBoosting methods direct + 8 threads", {
forecast_params <- list()
forecast_params$m <- 3
forecast_params$forecasting_params <- list()
forecast_params$forecasting_params$n_threads <- 2
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - PCA + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - PCA + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Base autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "base"
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Base Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Base autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "base"
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Base Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep"
dim_params$deep_layers <- c(10,5,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep"
dim_params$deep_layers <- c(10,5,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - LSTM autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "lstm"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$method <- forecasting_method
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - LSTM autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "lstm"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$method <- forecasting_method
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep LSTM autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_lstm"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep LSTM autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_lstm"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - GRU autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "gru"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - GRU autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "gru"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep GRU autoencoder + gradientBoosting direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_gru"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep GRU autoencoder + gradientBoosting recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_gru"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
jdestefani/ExtendedDFML documentation built on Dec. 20, 2021, 10:04 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context("DFML - Test gradient boosting")
library(keras)
library(dplyr)
library(forecast)
library(parallel)
library(xgboost)
library(lightgbm)
library(MEMTS)
#Set up - Data
X <- read.csv("testdata/Sigma4.ssv",sep=" ",header = TRUE)
X <- as.matrix(X)
splitting_point <- round(2*nrow(X)/3)
X_train <- scale(X[1:splitting_point,])
components <- 3
horizon <- 5
time_window <- horizon
ss_results.df <- data.frame(DimensionalityMethod=character(),ForecastingMethod=character(),Dataset=character(),Horizon=numeric(),Columns=numeric(),Time=numeric(),MSE=numeric(),Samples=numeric(),stringsAsFactors = FALSE)
# If there is an error with the Arima package, check
#detach("package:dse", unload=TRUE)
test_that("[DFML] - PCA + gradientBoosting methods direct + 8 threads", {
forecast_params <- list()
forecast_params$m <- 3
forecast_params$forecasting_params <- list()
forecast_params$forecasting_params$n_threads <- 2
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - PCA + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - PCA + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"PCA",
"gradientBoosting",
dimensionality_parameters = NULL,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="PCA",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Base autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "base"
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Base Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Base autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "base"
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Base Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep"
dim_params$deep_layers <- c(10,5,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep"
dim_params$deep_layers <- c(10,5,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - LSTM autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "lstm"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$method <- forecasting_method
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - LSTM autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "lstm"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$method <- forecasting_method
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep LSTM autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_lstm"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep LSTM autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_lstm"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep LSTM Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - GRU autoencoder + gradientBoosting methods direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "gru"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - GRU autoencoder + gradientBoosting methods recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "gru"
dim_params$time_window <- time_window
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep GRU autoencoder + gradientBoosting direct", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_gru"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "direct"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
test_that("[DFML] - Deep GRU autoencoder + gradientBoosting recursive", {
forecast_params <- list()
forecast_params$m <- 3
dim_params <- list()
dim_params$method <- "deep_gru"
dim_params$time_window <- time_window
dim_params$deep_layers <- c(10,3)
dim_params$epochs <- 10
multistep_method <- "recursive"
for (forecasting_method in ExtendedDFML::GRADIENT_BOOSTING_METHODS) {
print(paste("[INFO] - Testing",forecasting_method,"- h:",horizon))
forecast_params$forecasting_method <- forecasting_method
forecast_params$multistep_method <- multistep_method
results <- ExtendedDFML::DFML(X_train,
"Autoencoder_Keras",
"gradientBoosting",
dimensionality_parameters = dim_params,
forecast_params,
components,
horizon)
MSE_forecast <- MMSE(X[(splitting_point+1):(splitting_point+horizon),],results$X_hat)
ss_results.df <- bind_rows(ss_results.df,
data.frame(DimensionalityMethod="Deep GRU Autoencoder",
ForecastingMethod=paste(forecasting_method,multistep_method,sep="_"),
Dataset="Sigma 4",
Horizon=as.numeric(horizon),
Columns=as.numeric(ncol(X)),
Time=as.numeric(results$Time_dim+results$Time_forecast), # Elapsed time
MSE=as.numeric(MSE_forecast$mean),
Samples=as.numeric(splitting_point)))
}
print(ss_results.df)
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.