evaluate: Evaluate a Model

In autokeras: R Interface to 'AutoKeras'

Description

Evaluate the best model for the given data.

Usage

## S3 method for class 'AutokerasModel'
evaluate(object, x_test, y_test = NULL, batch_size = 32, ...)

Arguments

object	: A trained AutokerasModel instance.
x_test	: Any allowed types according to the input node. Testing data. Check corresponding AutokerasModel help to note how it should be provided.
y_test	: Any allowed types according to the input node. Testing data. Check corresponding AutokerasModel help to note how it should be provided. Defaults to 'NULL'.
batch_size	: numeric. Defaults to '32'.
...	: Unused.

Value

numeric test loss (if the model has a single output and no metrics) or list of scalars (if the model has multiple outputs and/or metrics). The attribute model$metrics_names will give you the display labels for the scalar outputs.

Examples

## Not run: 
library("keras")

# use the MNIST dataset as an example
mnist <- dataset_mnist()
c(x_train, y_train) %<-% mnist$train
c(x_test, y_test) %<-% mnist$test

library("autokeras")

# Initialize the image classifier
clf <- model_image_classifier(max_trials = 10) %>% # It tries 10 different models
  fit(x_train, y_train) # Feed the image classifier with training data

# Predict with the best model
(predicted_y <- clf %>% predict(x_test))

# Evaluate the best model with testing data
clf %>% evaluate(x_test, y_test)

# Get the best trained Keras model, to work with the keras R library
export_model(clf)

## End(Not run)

autokeras documentation built on Jan. 17, 2021, 9:06 a.m.