2022-05-09--conda-test.R

#!/usr/bin/env Rscript
library(reticulate)
library(keras)

reticulate::use_condaenv(condaenv = 'pyrstudio')
#Keras::install_keras(method='conda', conda='pyrstudio'


mnist <- dataset_mnist()
x_train <- mnist$train$x
y_train <- mnist$train$y
x_test <- mnist$test$x
y_test <- mnist$test$y

x_train <- array_reshape(x_train, c(nrow(x_train), 784))
x_train <- x_train / 255

x_test <- array_reshape(x_test, c(nrow(x_test), 784))
x_test <- x_test / 255


y_train <- to_categorical(y_train, 10)
y_test <- to_categorical(y_test, 10)

model <- keras_model_sequential()

model |>
  layer_dense(units = 256, activation = 'gelu', input_shape = c(784)) |>
  layer_dropout(rate = 0.4) |>
  layer_dense(units = 128, activation = 'gelu')  |>
  layer_dropout(rate = 0.3) |>
  layer_dense(units = 10, activation = 'softmax')

summary(model)

model |>
  compile(
    loss = 'categorical_crossentropy',
    optimizer = optimizer_rmsprop(),
    metrics = c('accuracy'))

history <-
  model |>
    fit(
      x_train, y_train,
      epochs = 30, batch_size = 128,
      validation_split = 0.2
    )


plot(history)

model %>% evaluate(x_test, y_test)