Deep neural networks- regression
Daniel
2022-05-20
Deep neural networks for regression
Loading packages and data sets
library(readr)
library(keras)
library(plotly)data("Boston", package = "MASS")
data.set <- Bostondim(data.set)## [1] 506 14Convert dataframe to matrix without dimnames
library(DT)
# Cast dataframe as a matrix
data.set <- as.matrix(data.set)
# Remove column names
dimnames(data.set) = NULL
head(data.set)## [,1] [,2] [,3] [,4] [,5] [,6] [,7] [,8] [,9] [,10] [,11] [,12]
## [1,] 0.00632 18 2.31 0 0.538 6.575 65.2 4.0900 1 296 15.3 396.90
## [2,] 0.02731 0 7.07 0 0.469 6.421 78.9 4.9671 2 242 17.8 396.90
## [3,] 0.02729 0 7.07 0 0.469 7.185 61.1 4.9671 2 242 17.8 392.83
## [4,] 0.03237 0 2.18 0 0.458 6.998 45.8 6.0622 3 222 18.7 394.63
## [5,] 0.06905 0 2.18 0 0.458 7.147 54.2 6.0622 3 222 18.7 396.90
## [6,] 0.02985 0 2.18 0 0.458 6.430 58.7 6.0622 3 222 18.7 394.12
## [,13] [,14]
## [1,] 4.98 24.0
## [2,] 9.14 21.6
## [3,] 4.03 34.7
## [4,] 2.94 33.4
## [5,] 5.33 36.2
## [6,] 5.21 28.7summary(data.set[, 14])## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.00 17.02 21.20 22.53 25.00 50.00 hist( data.set[, 14])
Fig 1 Histogram of the target variable
Spiting training and test data
# Split for train and test data
set.seed(123)
indx <- sample(2,
nrow(data.set),
replace = TRUE,
prob = c(0.75, 0.25)) # Makes index with values 1 and 2x_train <- data.set[indx == 1, 1:13]
x_test <- data.set[indx == 2, 1:13]
y_train <- data.set[indx == 1, 14]
y_test <- data.set[indx == 2, 14]Normalizing xtrain and xtest data
x_train <- scale(x_train)
x_test <- scale(x_test)Creating the model
model <- keras_model_sequential() %>%
layer_dense(units = 25,
activation = "relu",
input_shape = c(13)) %>%
layer_dropout(0.2) %>%
layer_dense(units = 25,
activation = "relu") %>%
layer_dropout(0.2) %>%
layer_dense(units = 25,
activation = "relu") %>%
layer_dropout(0.2) %>%
layer_dense(units = 1)## Loaded Tensorflow version 2.8.0model %>% summary()## Model: "sequential"
## ________________________________________________________________________________
## Layer (type) Output Shape Param #
## ================================================================================
## dense_3 (Dense) (None, 25) 350
## dropout_2 (Dropout) (None, 25) 0
## dense_2 (Dense) (None, 25) 650
## dropout_1 (Dropout) (None, 25) 0
## dense_1 (Dense) (None, 25) 650
## dropout (Dropout) (None, 25) 0
## dense (Dense) (None, 1) 26
## ================================================================================
## Total params: 1,676
## Trainable params: 1,676
## Non-trainable params: 0
## ________________________________________________________________________________model %>% get_config()## {'name': 'sequential', 'layers': [{'class_name': 'InputLayer', 'config': {'batch_input_shape': (None, 13), 'dtype': 'float32', 'sparse': False, 'ragged': False, 'name': 'dense_3_input'}}, {'class_name': 'Dense', 'config': {'name': 'dense_3', 'trainable': True, 'batch_input_shape': (None, 13), 'dtype': 'float32', 'units': 25, 'activation': 'relu', 'use_bias': True, 'kernel_initializer': {'class_name': 'GlorotUniform', 'config': {'seed': None}}, 'bias_initializer': {'class_name': 'Zeros', 'config': {}}, 'kernel_regularizer': None, 'bias_regularizer': None, 'activity_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}}, {'class_name': 'Dropout', 'config': {'name': 'dropout_2', 'trainable': True, 'dtype': 'float32', 'rate': 0.2, 'noise_shape': None, 'seed': None}}, {'class_name': 'Dense', 'config': {'name': 'dense_2', 'trainable': True, 'dtype': 'float32', 'units': 25, 'activation': 'relu', 'use_bias': True, 'kernel_initializer': {'class_name': 'GlorotUniform', 'config': {'seed': None}}, 'bias_initializer': {'class_name': 'Zeros', 'config': {}}, 'kernel_regularizer': None, 'bias_regularizer': None, 'activity_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}}, {'class_name': 'Dropout', 'config': {'name': 'dropout_1', 'trainable': True, 'dtype': 'float32', 'rate': 0.2, 'noise_shape': None, 'seed': None}}, {'class_name': 'Dense', 'config': {'name': 'dense_1', 'trainable': True, 'dtype': 'float32', 'units': 25, 'activation': 'relu', 'use_bias': True, 'kernel_initializer': {'class_name': 'GlorotUniform', 'config': {'seed': None}}, 'bias_initializer': {'class_name': 'Zeros', 'config': {}}, 'kernel_regularizer': None, 'bias_regularizer': None, 'activity_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}}, {'class_name': 'Dropout', 'config': {'name': 'dropout', 'trainable': True, 'dtype': 'float32', 'rate': 0.2, 'noise_shape': None, 'seed': None}}, {'class_name': 'Dense', 'config': {'name': 'dense', 'trainable': True, 'dtype': 'float32', 'units': 1, 'activation': 'linear', 'use_bias': True, 'kernel_initializer': {'class_name': 'GlorotUniform', 'config': {'seed': None}}, 'bias_initializer': {'class_name': 'Zeros', 'config': {}}, 'kernel_regularizer': None, 'bias_regularizer': None, 'activity_regularizer': None, 'kernel_constraint': None, 'bias_constraint': None}}]}Compiling the model
model %>% compile(loss = "mse",
optimizer = optimizer_rmsprop(),
metrics = c("mean_absolute_error"))Fitting the model
history <- model %>%
fit(x_train,
y_train,
epoch = 100,
batch_size = 64,
validation_split = 0.1,
callbacks = c(callback_early_stopping(monitor = "val_mean_absolute_error",
patience = 5)),
verbose = 2)c(loss, mae) %<-% (model %>% evaluate(x_test, y_test, verbose = 0))
paste0("Mean absolute error on test set: ", sprintf("%.2f", mae))## [1] "Mean absolute error on test set: 3.88"Plot the training process
plot(history)
### Calculating the predicted values on test data
pred2 <- model %>%
predict(x_test) %>% k_get_value()
head(cbind(pred2,y_test))## y_test
## [1,] 25.08657 21.6
## [2,] 34.35121 33.4
## [3,] 34.24675 36.2
## [4,] 15.04251 27.1
## [5,] 15.59313 15.0
## [6,] 15.91344 19.9- calculating
mean absolute error and root mean square errorand ploting
error <- y_test-pred2
head(error)## [,1]
## [1,] -3.4865746
## [2,] -0.9512077
## [3,] 1.9532463
## [4,] 12.0574862
## [5,] -0.5931273
## [6,] 3.9865622rmse <- sqrt(mean(error)^2)
rmse## [1] 0.6797003plot(error)