Update: The original code has been updated to use the tidymodels init_split() function, rather than using the indicies method which originally used setdiff, which now may have a conflict between base R and the tidyverse.

Thank you

A big thank you to Leon Jessen for posting his code on github.

Building a simple neural network using Keras and Tensorflow

I have forked his project on github and put his code into an R Notebook so we can run it in class.

Motivation

The following is a minimal example for building your first simple artificial neural network using Keras and TensorFlow for R.

TensorFlow for R by Rstudio lives here.

Gettings started - Install Keras and TensorFlow for R

You can install the Keras for R package from CRAN as follows:

# install.packages("keras")

TensorFlow is the default backend engine. TensorFlow and Keras can be installed as follows:

# library(keras)
# install_keras()

Naturally, we will also need Tidyverse.

# Install from CRAN
# install.packages("tidyverse")

# Or the development version from GitHub
# install.packages("devtools")
# devtools::install_github("hadley/tidyverse")

Once installed, we simply load the libraries.

library("keras")
suppressMessages(library("tidyverse"))

Artificial Neural Network Using the Iris Data Set

Right, let’s get to it!

Data

The famous (Fisher’s or Anderson’s) iris data set contains a total of 150 observations of 4 input features Sepal.Length, Sepal.Width, Petal.Length and Petal.Width and 3 output classes setosa versicolor and virginica, with 50 observations in each class. The distributions of the feature values looks like so:

iris_tib <- as_tibble(iris)
iris_tib
iris_tib %>% pivot_longer(names_to = "feature", values_to = "value", -Species) %>%
  ggplot(aes(x = feature, y = value, fill = Species)) +
  geom_violin(alpha = 0.5, scale = "width") +
  theme_bw()

Our aim is to connect the 4 input features to the correct output class using an artificial neural network. For this task, we have chosen the following simple architecture with one input layer with 4 neurons (one for each feature), one hidden layer with 4 neurons and one output layer with 3 neurons (one for each class), all fully connected.

architecture_visualisation.png

Our artificial neural network will have a total of 35 parameters: 4 for each input neuron connected to the hidden layer, plus an additional 4 for the associated first bias neuron and 3 for each of the hidden neurons connected to the output layer, plus an additional 3 for the associated second bias neuron, i.e. \(4 \times 4 + 4 + 4 \times 3 + 3=35\)

Prepare data

We start with slightly wrangling the iris data set by renaming and scaling the features and converting character labels to numeric.

set.seed(265509)
nn_dat <- iris_tib %>%
  mutate(sepal_length = scale(Sepal.Length),
         sepal_width  = scale(Sepal.Width),
         petal_length = scale(Petal.Length),
         petal_width  = scale(Petal.Width),          
         class_label  = as.numeric(Species) - 1) %>% 
    select(sepal_length, sepal_width, petal_length, petal_width, class_label)

nn_dat %>% head()

Then, we create indices for splitting the iris data into a training and a test data set. We set aside 20% of the data for testing.

library(tidymodels)

set.seed(364)
n <- nrow(nn_dat)
n
[1] 150
iris_parts <- nn_dat %>%
  initial_split(prop = 0.8)

train <- iris_parts %>%
  training()

test <- iris_parts %>%
  testing()

list(train, test) %>%
  map_int(nrow)
[1] 121  29

n_total_samples <- nrow(nn_dat)

n_train_samples <- nrow(train)

n_test_samples <- nrow(test)

Create training and test data

Note that the functions in the keras package are expecting the data to be in a matrix object and not a tibble. So as.matrix is added at the end of each line.

x_train <- train %>% select(-class_label) %>% as.matrix()
y_train <- train %>% select(class_label) %>% as.matrix() %>% to_categorical()

x_test <- test %>% select(-class_label) %>% as.matrix()
y_test <- test %>% select(class_label) %>% as.matrix() %>% to_categorical() 

dim(y_train)
[1] 121   3
dim(y_test)
[1] 29  3

Set Architecture

With the data in place, we now set the architecture of our neural network.

model <- keras_model_sequential()
model %>% 
  layer_dense(units = 4, activation = 'relu', input_shape = 4) %>% 
  layer_dense(units = 3, activation = 'softmax')
model %>% summary
Model: "sequential_3"
____________________________________________________________________________
Layer (type)                      Output Shape                  Param #     
============================================================================
dense_7 (Dense)                   (None, 4)                     20          
____________________________________________________________________________
dense_6 (Dense)                   (None, 3)                     15          
============================================================================
Total params: 35
Trainable params: 35
Non-trainable params: 0
____________________________________________________________________________

Next, the architecture set in the model needs to be compiled.

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

Train the Artificial Neural Network

Lastly we fit the model and save the training progress in the history object.

Try changing the validation_split from 0 to 0.2 to see the validation_loss.

history <- model %>% fit(
  x = x_train, y = y_train,
  epochs = 200,
  batch_size = 20,
  validation_split = 0.2
)
Epoch 1/200

1/5 [=====>........................] - ETA: 0s - loss: 1.3388 - accuracy: 0.3000
5/5 [==============================] - 0s 626us/step - loss: 1.2012 - accuracy: 0.3958

5/5 [==============================] - 1s 116ms/step - loss: 1.2012 - accuracy: 0.3958 - val_loss: 1.8751 - val_accuracy: 0.0000e+00
Epoch 2/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0707 - accuracy: 0.4500
5/5 [==============================] - 0s 552us/step - loss: 1.1734 - accuracy: 0.3958

5/5 [==============================] - 0s 39ms/step - loss: 1.1734 - accuracy: 0.3958 - val_loss: 1.8398 - val_accuracy: 0.0000e+00
Epoch 3/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0602 - accuracy: 0.4500
5/5 [==============================] - 0s 3ms/step - loss: 1.1548 - accuracy: 0.3958

5/5 [==============================] - 0s 33ms/step - loss: 1.1548 - accuracy: 0.3958 - val_loss: 1.8107 - val_accuracy: 0.0000e+00
Epoch 4/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0990 - accuracy: 0.5000
5/5 [==============================] - 0s 512us/step - loss: 1.1382 - accuracy: 0.3958

5/5 [==============================] - 0s 26ms/step - loss: 1.1382 - accuracy: 0.3958 - val_loss: 1.7815 - val_accuracy: 0.0000e+00
Epoch 5/200

1/5 [=====>........................] - ETA: 0s - loss: 1.2690 - accuracy: 0.3000
5/5 [==============================] - 0s 2ms/step - loss: 1.1236 - accuracy: 0.3958

5/5 [==============================] - 0s 32ms/step - loss: 1.1236 - accuracy: 0.3958 - val_loss: 1.7558 - val_accuracy: 0.0000e+00
Epoch 6/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8600 - accuracy: 0.6000
5/5 [==============================] - 0s 2ms/step - loss: 1.1097 - accuracy: 0.3958

5/5 [==============================] - 0s 31ms/step - loss: 1.1097 - accuracy: 0.3958 - val_loss: 1.7269 - val_accuracy: 0.0000e+00
Epoch 7/200

1/5 [=====>........................] - ETA: 0s - loss: 1.1558 - accuracy: 0.3000
5/5 [==============================] - 0s 569us/step - loss: 1.0960 - accuracy: 0.3958

5/5 [==============================] - 0s 26ms/step - loss: 1.0960 - accuracy: 0.3958 - val_loss: 1.7008 - val_accuracy: 0.0000e+00
Epoch 8/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0330 - accuracy: 0.4500
5/5 [==============================] - 0s 697us/step - loss: 1.0831 - accuracy: 0.3958

5/5 [==============================] - 0s 27ms/step - loss: 1.0831 - accuracy: 0.3958 - val_loss: 1.6750 - val_accuracy: 0.0000e+00
Epoch 9/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0609 - accuracy: 0.3500
5/5 [==============================] - 0s 3ms/step - loss: 1.0705 - accuracy: 0.4062

5/5 [==============================] - 0s 32ms/step - loss: 1.0705 - accuracy: 0.4062 - val_loss: 1.6506 - val_accuracy: 0.0000e+00
Epoch 10/200

1/5 [=====>........................] - ETA: 0s - loss: 1.1090 - accuracy: 0.4500
5/5 [==============================] - 0s 800us/step - loss: 1.0585 - accuracy: 0.4062

5/5 [==============================] - 0s 29ms/step - loss: 1.0585 - accuracy: 0.4062 - val_loss: 1.6273 - val_accuracy: 0.0000e+00
Epoch 11/200

1/5 [=====>........................] - ETA: 0s - loss: 1.1656 - accuracy: 0.4000
5/5 [==============================] - 0s 3ms/step - loss: 1.0471 - accuracy: 0.4167

5/5 [==============================] - 0s 34ms/step - loss: 1.0471 - accuracy: 0.4167 - val_loss: 1.6086 - val_accuracy: 0.0000e+00
Epoch 12/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9748 - accuracy: 0.4000
5/5 [==============================] - 0s 3ms/step - loss: 1.0362 - accuracy: 0.4167

5/5 [==============================] - 0s 30ms/step - loss: 1.0362 - accuracy: 0.4167 - val_loss: 1.5883 - val_accuracy: 0.0000e+00
Epoch 13/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8675 - accuracy: 0.5500
5/5 [==============================] - 0s 3ms/step - loss: 1.0256 - accuracy: 0.4479

5/5 [==============================] - 0s 32ms/step - loss: 1.0256 - accuracy: 0.4479 - val_loss: 1.5667 - val_accuracy: 0.0000e+00
Epoch 14/200

1/5 [=====>........................] - ETA: 0s - loss: 1.2064 - accuracy: 0.3000
5/5 [==============================] - 0s 630us/step - loss: 1.0151 - accuracy: 0.4583

5/5 [==============================] - 0s 26ms/step - loss: 1.0151 - accuracy: 0.4583 - val_loss: 1.5503 - val_accuracy: 0.0000e+00
Epoch 15/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9733 - accuracy: 0.5000
5/5 [==============================] - 0s 642us/step - loss: 1.0052 - accuracy: 0.4688

5/5 [==============================] - 0s 26ms/step - loss: 1.0052 - accuracy: 0.4688 - val_loss: 1.5308 - val_accuracy: 0.0000e+00
Epoch 16/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8383 - accuracy: 0.6000
5/5 [==============================] - 0s 604us/step - loss: 0.9959 - accuracy: 0.4792

5/5 [==============================] - 0s 26ms/step - loss: 0.9959 - accuracy: 0.4792 - val_loss: 1.5137 - val_accuracy: 0.0000e+00
Epoch 17/200

1/5 [=====>........................] - ETA: 0s - loss: 1.1821 - accuracy: 0.3500
5/5 [==============================] - 0s 2ms/step - loss: 0.9867 - accuracy: 0.4896

5/5 [==============================] - 0s 31ms/step - loss: 0.9867 - accuracy: 0.4896 - val_loss: 1.4975 - val_accuracy: 0.0000e+00
Epoch 18/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0665 - accuracy: 0.5500
5/5 [==============================] - 0s 3ms/step - loss: 0.9785 - accuracy: 0.5625

5/5 [==============================] - 0s 32ms/step - loss: 0.9785 - accuracy: 0.5625 - val_loss: 1.4783 - val_accuracy: 0.0000e+00
Epoch 19/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9199 - accuracy: 0.5500
5/5 [==============================] - 0s 2ms/step - loss: 0.9698 - accuracy: 0.5833

5/5 [==============================] - 0s 33ms/step - loss: 0.9698 - accuracy: 0.5833 - val_loss: 1.4582 - val_accuracy: 0.0000e+00
Epoch 20/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8482 - accuracy: 0.7000
5/5 [==============================] - 0s 693us/step - loss: 0.9616 - accuracy: 0.6146

5/5 [==============================] - 0s 26ms/step - loss: 0.9616 - accuracy: 0.6146 - val_loss: 1.4403 - val_accuracy: 0.0000e+00
Epoch 21/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8360 - accuracy: 0.8000
5/5 [==============================] - 0s 609us/step - loss: 0.9538 - accuracy: 0.6146

5/5 [==============================] - 0s 28ms/step - loss: 0.9538 - accuracy: 0.6146 - val_loss: 1.4225 - val_accuracy: 0.0000e+00
Epoch 22/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9203 - accuracy: 0.6000
5/5 [==============================] - 0s 1ms/step - loss: 0.9457 - accuracy: 0.6354

5/5 [==============================] - 0s 29ms/step - loss: 0.9457 - accuracy: 0.6354 - val_loss: 1.4073 - val_accuracy: 0.0000e+00
Epoch 23/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8758 - accuracy: 0.7500
5/5 [==============================] - 0s 2ms/step - loss: 0.9382 - accuracy: 0.6771

5/5 [==============================] - 0s 31ms/step - loss: 0.9382 - accuracy: 0.6771 - val_loss: 1.3912 - val_accuracy: 0.0000e+00
Epoch 24/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9551 - accuracy: 0.7000
5/5 [==============================] - 0s 2ms/step - loss: 0.9309 - accuracy: 0.6875

5/5 [==============================] - 0s 31ms/step - loss: 0.9309 - accuracy: 0.6875 - val_loss: 1.3763 - val_accuracy: 0.0000e+00
Epoch 25/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8751 - accuracy: 0.6500
5/5 [==============================] - 0s 2ms/step - loss: 0.9234 - accuracy: 0.6979

5/5 [==============================] - 0s 32ms/step - loss: 0.9234 - accuracy: 0.6979 - val_loss: 1.3585 - val_accuracy: 0.0000e+00
Epoch 26/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9436 - accuracy: 0.7500
5/5 [==============================] - 0s 2ms/step - loss: 0.9160 - accuracy: 0.6979

5/5 [==============================] - 0s 31ms/step - loss: 0.9160 - accuracy: 0.6979 - val_loss: 1.3445 - val_accuracy: 0.0000e+00
Epoch 27/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9072 - accuracy: 0.7000
5/5 [==============================] - 0s 560us/step - loss: 0.9085 - accuracy: 0.6979

5/5 [==============================] - 0s 28ms/step - loss: 0.9085 - accuracy: 0.6979 - val_loss: 1.3290 - val_accuracy: 0.0000e+00
Epoch 28/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9360 - accuracy: 0.6000
5/5 [==============================] - 0s 538us/step - loss: 0.9012 - accuracy: 0.6979

5/5 [==============================] - 0s 26ms/step - loss: 0.9012 - accuracy: 0.6979 - val_loss: 1.3149 - val_accuracy: 0.0000e+00
Epoch 29/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8873 - accuracy: 0.8000
5/5 [==============================] - 0s 606us/step - loss: 0.8936 - accuracy: 0.6979

5/5 [==============================] - 0s 27ms/step - loss: 0.8936 - accuracy: 0.6979 - val_loss: 1.3004 - val_accuracy: 0.0000e+00
Epoch 30/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7758 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.8860 - accuracy: 0.7188

5/5 [==============================] - 0s 33ms/step - loss: 0.8860 - accuracy: 0.7188 - val_loss: 1.2868 - val_accuracy: 0.0000e+00
Epoch 31/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7663 - accuracy: 0.9000
5/5 [==============================] - 0s 3ms/step - loss: 0.8789 - accuracy: 0.7292

5/5 [==============================] - 0s 34ms/step - loss: 0.8789 - accuracy: 0.7292 - val_loss: 1.2715 - val_accuracy: 0.0000e+00
Epoch 32/200

1/5 [=====>........................] - ETA: 0s - loss: 1.0765 - accuracy: 0.5000
5/5 [==============================] - 0s 3ms/step - loss: 0.8718 - accuracy: 0.7292

5/5 [==============================] - 0s 33ms/step - loss: 0.8718 - accuracy: 0.7292 - val_loss: 1.2613 - val_accuracy: 0.0000e+00
Epoch 33/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8815 - accuracy: 0.7500
5/5 [==============================] - 0s 607us/step - loss: 0.8643 - accuracy: 0.7396

5/5 [==============================] - 0s 26ms/step - loss: 0.8643 - accuracy: 0.7396 - val_loss: 1.2473 - val_accuracy: 0.0000e+00
Epoch 34/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7963 - accuracy: 0.8000
5/5 [==============================] - 0s 849us/step - loss: 0.8573 - accuracy: 0.7604

5/5 [==============================] - 0s 27ms/step - loss: 0.8573 - accuracy: 0.7604 - val_loss: 1.2358 - val_accuracy: 0.0000e+00
Epoch 35/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7933 - accuracy: 0.8000
5/5 [==============================] - 0s 612us/step - loss: 0.8495 - accuracy: 0.7604

5/5 [==============================] - 0s 26ms/step - loss: 0.8495 - accuracy: 0.7604 - val_loss: 1.2218 - val_accuracy: 0.0000e+00
Epoch 36/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8034 - accuracy: 0.8000
5/5 [==============================] - 0s 3ms/step - loss: 0.8423 - accuracy: 0.7604

5/5 [==============================] - 0s 33ms/step - loss: 0.8423 - accuracy: 0.7604 - val_loss: 1.2092 - val_accuracy: 0.0000e+00
Epoch 37/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9345 - accuracy: 0.7000
5/5 [==============================] - 0s 2ms/step - loss: 0.8353 - accuracy: 0.7604

5/5 [==============================] - 0s 32ms/step - loss: 0.8353 - accuracy: 0.7604 - val_loss: 1.2018 - val_accuracy: 0.0000e+00
Epoch 38/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7590 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.8283 - accuracy: 0.7604

5/5 [==============================] - 0s 32ms/step - loss: 0.8283 - accuracy: 0.7604 - val_loss: 1.1932 - val_accuracy: 0.0000e+00
Epoch 39/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7568 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.8209 - accuracy: 0.7708

5/5 [==============================] - 0s 33ms/step - loss: 0.8209 - accuracy: 0.7708 - val_loss: 1.1809 - val_accuracy: 0.0000e+00
Epoch 40/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8633 - accuracy: 0.7500
5/5 [==============================] - 0s 523us/step - loss: 0.8134 - accuracy: 0.7812

5/5 [==============================] - 0s 27ms/step - loss: 0.8134 - accuracy: 0.7812 - val_loss: 1.1705 - val_accuracy: 0.0000e+00
Epoch 41/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8138 - accuracy: 0.7500
5/5 [==============================] - 0s 663us/step - loss: 0.8060 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.8060 - accuracy: 0.7917 - val_loss: 1.1600 - val_accuracy: 0.0000e+00
Epoch 42/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8629 - accuracy: 0.7000
5/5 [==============================] - 0s 3ms/step - loss: 0.7989 - accuracy: 0.7917

5/5 [==============================] - 0s 35ms/step - loss: 0.7989 - accuracy: 0.7917 - val_loss: 1.1503 - val_accuracy: 0.0000e+00
Epoch 43/200

1/5 [=====>........................] - ETA: 0s - loss: 0.9243 - accuracy: 0.7000
5/5 [==============================] - 0s 2ms/step - loss: 0.7914 - accuracy: 0.7917

5/5 [==============================] - 0s 31ms/step - loss: 0.7914 - accuracy: 0.7917 - val_loss: 1.1406 - val_accuracy: 0.0000e+00
Epoch 44/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7645 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.7841 - accuracy: 0.7917

5/5 [==============================] - 0s 33ms/step - loss: 0.7841 - accuracy: 0.7917 - val_loss: 1.1321 - val_accuracy: 0.0000e+00
Epoch 45/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7493 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.7766 - accuracy: 0.7917

5/5 [==============================] - 0s 32ms/step - loss: 0.7766 - accuracy: 0.7917 - val_loss: 1.1219 - val_accuracy: 0.0000e+00
Epoch 46/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8219 - accuracy: 0.7000
5/5 [==============================] - 0s 645us/step - loss: 0.7690 - accuracy: 0.7917

5/5 [==============================] - 0s 27ms/step - loss: 0.7690 - accuracy: 0.7917 - val_loss: 1.1113 - val_accuracy: 0.0000e+00
Epoch 47/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7436 - accuracy: 0.7500
5/5 [==============================] - 0s 524us/step - loss: 0.7616 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.7616 - accuracy: 0.7917 - val_loss: 1.1004 - val_accuracy: 0.0000e+00
Epoch 48/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8330 - accuracy: 0.6500
5/5 [==============================] - 0s 2ms/step - loss: 0.7543 - accuracy: 0.7917

5/5 [==============================] - 0s 32ms/step - loss: 0.7543 - accuracy: 0.7917 - val_loss: 1.0938 - val_accuracy: 0.0000e+00
Epoch 49/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7690 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.7469 - accuracy: 0.7917

5/5 [==============================] - 0s 33ms/step - loss: 0.7469 - accuracy: 0.7917 - val_loss: 1.0851 - val_accuracy: 0.0000e+00
Epoch 50/200

1/5 [=====>........................] - ETA: 0s - loss: 0.8127 - accuracy: 0.7500
5/5 [==============================] - 0s 3ms/step - loss: 0.7395 - accuracy: 0.7917

5/5 [==============================] - 0s 34ms/step - loss: 0.7395 - accuracy: 0.7917 - val_loss: 1.0767 - val_accuracy: 0.0000e+00
Epoch 51/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7547 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.7322 - accuracy: 0.7917

5/5 [==============================] - 0s 31ms/step - loss: 0.7322 - accuracy: 0.7917 - val_loss: 1.0657 - val_accuracy: 0.0000e+00
Epoch 52/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7456 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.7248 - accuracy: 0.7917

5/5 [==============================] - 0s 29ms/step - loss: 0.7248 - accuracy: 0.7917 - val_loss: 1.0564 - val_accuracy: 0.0000e+00
Epoch 53/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6830 - accuracy: 0.8000
5/5 [==============================] - 0s 606us/step - loss: 0.7176 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.7176 - accuracy: 0.7917 - val_loss: 1.0487 - val_accuracy: 0.0000e+00
Epoch 54/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6858 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.7105 - accuracy: 0.7917

5/5 [==============================] - 0s 31ms/step - loss: 0.7105 - accuracy: 0.7917 - val_loss: 1.0416 - val_accuracy: 0.0000e+00
Epoch 55/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7632 - accuracy: 0.7000
5/5 [==============================] - 0s 2ms/step - loss: 0.7038 - accuracy: 0.7917

5/5 [==============================] - 0s 32ms/step - loss: 0.7038 - accuracy: 0.7917 - val_loss: 1.0348 - val_accuracy: 0.0000e+00
Epoch 56/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7497 - accuracy: 0.7500
5/5 [==============================] - 0s 3ms/step - loss: 0.6965 - accuracy: 0.7917

5/5 [==============================] - 0s 68ms/step - loss: 0.6965 - accuracy: 0.7917 - val_loss: 1.0289 - val_accuracy: 0.0000e+00
Epoch 57/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7487 - accuracy: 0.6500
5/5 [==============================] - 0s 3ms/step - loss: 0.6898 - accuracy: 0.7917

5/5 [==============================] - 0s 33ms/step - loss: 0.6898 - accuracy: 0.7917 - val_loss: 1.0226 - val_accuracy: 0.0000e+00
Epoch 58/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7547 - accuracy: 0.8000
5/5 [==============================] - 0s 608us/step - loss: 0.6828 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.6828 - accuracy: 0.7917 - val_loss: 1.0144 - val_accuracy: 0.0000e+00
Epoch 59/200

1/5 [=====>........................] - ETA: 0s - loss: 0.7410 - accuracy: 0.7000
5/5 [==============================] - 0s 655us/step - loss: 0.6758 - accuracy: 0.7917

5/5 [==============================] - 0s 27ms/step - loss: 0.6758 - accuracy: 0.7917 - val_loss: 1.0091 - val_accuracy: 0.0000e+00
Epoch 60/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6408 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.6692 - accuracy: 0.7917

5/5 [==============================] - 0s 31ms/step - loss: 0.6692 - accuracy: 0.7917 - val_loss: 1.0021 - val_accuracy: 0.0000e+00
Epoch 61/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6549 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.6626 - accuracy: 0.7917

5/5 [==============================] - 0s 31ms/step - loss: 0.6626 - accuracy: 0.7917 - val_loss: 0.9957 - val_accuracy: 0.0000e+00
Epoch 62/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6856 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.6559 - accuracy: 0.7917

5/5 [==============================] - 0s 32ms/step - loss: 0.6559 - accuracy: 0.7917 - val_loss: 0.9879 - val_accuracy: 0.0000e+00
Epoch 63/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6367 - accuracy: 0.8500
5/5 [==============================] - 0s 1ms/step - loss: 0.6494 - accuracy: 0.7917

5/5 [==============================] - 0s 30ms/step - loss: 0.6494 - accuracy: 0.7917 - val_loss: 0.9811 - val_accuracy: 0.0000e+00
Epoch 64/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6529 - accuracy: 0.7000
5/5 [==============================] - 0s 613us/step - loss: 0.6425 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.6425 - accuracy: 0.7917 - val_loss: 0.9743 - val_accuracy: 0.0000e+00
Epoch 65/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6007 - accuracy: 0.8500
5/5 [==============================] - 0s 663us/step - loss: 0.6362 - accuracy: 0.7917

5/5 [==============================] - 0s 26ms/step - loss: 0.6362 - accuracy: 0.7917 - val_loss: 0.9675 - val_accuracy: 0.0000e+00
Epoch 66/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5578 - accuracy: 0.7500
5/5 [==============================] - 0s 3ms/step - loss: 0.6293 - accuracy: 0.7917

5/5 [==============================] - 0s 32ms/step - loss: 0.6293 - accuracy: 0.7917 - val_loss: 0.9620 - val_accuracy: 0.0000e+00
Epoch 67/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6102 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.6233 - accuracy: 0.7812

5/5 [==============================] - 0s 32ms/step - loss: 0.6233 - accuracy: 0.7812 - val_loss: 0.9557 - val_accuracy: 0.0000e+00
Epoch 68/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5422 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.6168 - accuracy: 0.7812

5/5 [==============================] - 0s 33ms/step - loss: 0.6168 - accuracy: 0.7812 - val_loss: 0.9473 - val_accuracy: 0.0000e+00
Epoch 69/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6157 - accuracy: 0.9500
5/5 [==============================] - 0s 3ms/step - loss: 0.6104 - accuracy: 0.7812

5/5 [==============================] - 0s 33ms/step - loss: 0.6104 - accuracy: 0.7812 - val_loss: 0.9386 - val_accuracy: 0.0000e+00
Epoch 70/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6297 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.6038 - accuracy: 0.7812

5/5 [==============================] - 0s 33ms/step - loss: 0.6038 - accuracy: 0.7812 - val_loss: 0.9315 - val_accuracy: 0.0000e+00
Epoch 71/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5232 - accuracy: 0.9000
5/5 [==============================] - 0s 551us/step - loss: 0.5976 - accuracy: 0.7812

5/5 [==============================] - 0s 26ms/step - loss: 0.5976 - accuracy: 0.7812 - val_loss: 0.9239 - val_accuracy: 0.0000e+00
Epoch 72/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5390 - accuracy: 0.8500
5/5 [==============================] - 0s 557us/step - loss: 0.5914 - accuracy: 0.7812

5/5 [==============================] - 0s 26ms/step - loss: 0.5914 - accuracy: 0.7812 - val_loss: 0.9178 - val_accuracy: 0.0000e+00
Epoch 73/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4926 - accuracy: 0.9000
5/5 [==============================] - 0s 3ms/step - loss: 0.5850 - accuracy: 0.8125

5/5 [==============================] - 0s 31ms/step - loss: 0.5850 - accuracy: 0.8125 - val_loss: 0.9114 - val_accuracy: 0.1200
Epoch 74/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5390 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.5788 - accuracy: 0.8125

5/5 [==============================] - 0s 32ms/step - loss: 0.5788 - accuracy: 0.8125 - val_loss: 0.9038 - val_accuracy: 0.3600
Epoch 75/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5202 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.5724 - accuracy: 0.8333

5/5 [==============================] - 0s 31ms/step - loss: 0.5724 - accuracy: 0.8333 - val_loss: 0.8982 - val_accuracy: 0.4000
Epoch 76/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5584 - accuracy: 0.7500
5/5 [==============================] - 0s 2ms/step - loss: 0.5663 - accuracy: 0.8646

5/5 [==============================] - 0s 32ms/step - loss: 0.5663 - accuracy: 0.8646 - val_loss: 0.8926 - val_accuracy: 0.4000
Epoch 77/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5535 - accuracy: 0.8500
5/5 [==============================] - 0s 626us/step - loss: 0.5601 - accuracy: 0.8646

5/5 [==============================] - 0s 27ms/step - loss: 0.5601 - accuracy: 0.8646 - val_loss: 0.8847 - val_accuracy: 0.4400
Epoch 78/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6309 - accuracy: 0.8500
5/5 [==============================] - 0s 594us/step - loss: 0.5539 - accuracy: 0.8646

5/5 [==============================] - 0s 26ms/step - loss: 0.5539 - accuracy: 0.8646 - val_loss: 0.8753 - val_accuracy: 0.4400
Epoch 79/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6024 - accuracy: 0.7500
5/5 [==============================] - 0s 537us/step - loss: 0.5477 - accuracy: 0.8646

5/5 [==============================] - 0s 27ms/step - loss: 0.5477 - accuracy: 0.8646 - val_loss: 0.8672 - val_accuracy: 0.4400
Epoch 80/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5014 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.5422 - accuracy: 0.8646

5/5 [==============================] - 0s 31ms/step - loss: 0.5422 - accuracy: 0.8646 - val_loss: 0.8594 - val_accuracy: 0.4400
Epoch 81/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6606 - accuracy: 0.7000
5/5 [==============================] - 0s 3ms/step - loss: 0.5361 - accuracy: 0.8646

5/5 [==============================] - 0s 33ms/step - loss: 0.5361 - accuracy: 0.8646 - val_loss: 0.8531 - val_accuracy: 0.4400
Epoch 82/200

1/5 [=====>........................] - ETA: 0s - loss: 0.6058 - accuracy: 0.7500
5/5 [==============================] - 0s 1ms/step - loss: 0.5302 - accuracy: 0.8750

5/5 [==============================] - 0s 31ms/step - loss: 0.5302 - accuracy: 0.8750 - val_loss: 0.8477 - val_accuracy: 0.4800
Epoch 83/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5446 - accuracy: 0.8000
5/5 [==============================] - 0s 1ms/step - loss: 0.5243 - accuracy: 0.8750

5/5 [==============================] - 0s 30ms/step - loss: 0.5243 - accuracy: 0.8750 - val_loss: 0.8401 - val_accuracy: 0.5200
Epoch 84/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5798 - accuracy: 0.8500
5/5 [==============================] - 0s 634us/step - loss: 0.5187 - accuracy: 0.8750

5/5 [==============================] - 0s 26ms/step - loss: 0.5187 - accuracy: 0.8750 - val_loss: 0.8311 - val_accuracy: 0.5200
Epoch 85/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4894 - accuracy: 0.9500
5/5 [==============================] - 0s 619us/step - loss: 0.5131 - accuracy: 0.8750

5/5 [==============================] - 0s 26ms/step - loss: 0.5131 - accuracy: 0.8750 - val_loss: 0.8218 - val_accuracy: 0.5200
Epoch 86/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5637 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.5075 - accuracy: 0.8750

5/5 [==============================] - 0s 32ms/step - loss: 0.5075 - accuracy: 0.8750 - val_loss: 0.8147 - val_accuracy: 0.5200
Epoch 87/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5724 - accuracy: 0.8000
5/5 [==============================] - 0s 2ms/step - loss: 0.5017 - accuracy: 0.8750

5/5 [==============================] - 0s 32ms/step - loss: 0.5017 - accuracy: 0.8750 - val_loss: 0.8067 - val_accuracy: 0.5200
Epoch 88/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4510 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.4963 - accuracy: 0.8750

5/5 [==============================] - 0s 33ms/step - loss: 0.4963 - accuracy: 0.8750 - val_loss: 0.7977 - val_accuracy: 0.5600
Epoch 89/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4442 - accuracy: 0.9000
5/5 [==============================] - 0s 632us/step - loss: 0.4905 - accuracy: 0.8750

5/5 [==============================] - 0s 26ms/step - loss: 0.4905 - accuracy: 0.8750 - val_loss: 0.7886 - val_accuracy: 0.5600
Epoch 90/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5058 - accuracy: 0.7500
5/5 [==============================] - 0s 3ms/step - loss: 0.4854 - accuracy: 0.8854

5/5 [==============================] - 0s 32ms/step - loss: 0.4854 - accuracy: 0.8854 - val_loss: 0.7831 - val_accuracy: 0.5600
Epoch 91/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4314 - accuracy: 0.9500
5/5 [==============================] - 0s 565us/step - loss: 0.4799 - accuracy: 0.8854

5/5 [==============================] - 0s 26ms/step - loss: 0.4799 - accuracy: 0.8854 - val_loss: 0.7738 - val_accuracy: 0.5600
Epoch 92/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4514 - accuracy: 0.9000
5/5 [==============================] - 0s 628us/step - loss: 0.4746 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4746 - accuracy: 0.8958 - val_loss: 0.7651 - val_accuracy: 0.6000
Epoch 93/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4436 - accuracy: 0.9000
5/5 [==============================] - 0s 605us/step - loss: 0.4696 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4696 - accuracy: 0.8958 - val_loss: 0.7561 - val_accuracy: 0.6400
Epoch 94/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4046 - accuracy: 0.9000
5/5 [==============================] - 0s 555us/step - loss: 0.4646 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4646 - accuracy: 0.8958 - val_loss: 0.7493 - val_accuracy: 0.6400
Epoch 95/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3983 - accuracy: 0.9000
5/5 [==============================] - 0s 611us/step - loss: 0.4597 - accuracy: 0.8958

5/5 [==============================] - 0s 27ms/step - loss: 0.4597 - accuracy: 0.8958 - val_loss: 0.7436 - val_accuracy: 0.6400
Epoch 96/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4426 - accuracy: 0.9500
5/5 [==============================] - 0s 630us/step - loss: 0.4552 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4552 - accuracy: 0.8958 - val_loss: 0.7362 - val_accuracy: 0.6400
Epoch 97/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4589 - accuracy: 0.8000
5/5 [==============================] - 0s 611us/step - loss: 0.4502 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4502 - accuracy: 0.8958 - val_loss: 0.7322 - val_accuracy: 0.6400
Epoch 98/200

1/5 [=====>........................] - ETA: 0s - loss: 0.5474 - accuracy: 0.8500
5/5 [==============================] - 0s 617us/step - loss: 0.4457 - accuracy: 0.8958

5/5 [==============================] - 0s 27ms/step - loss: 0.4457 - accuracy: 0.8958 - val_loss: 0.7259 - val_accuracy: 0.6400
Epoch 99/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4587 - accuracy: 0.8500
5/5 [==============================] - 0s 618us/step - loss: 0.4410 - accuracy: 0.8958

5/5 [==============================] - 0s 26ms/step - loss: 0.4410 - accuracy: 0.8958 - val_loss: 0.7210 - val_accuracy: 0.6400
Epoch 100/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3504 - accuracy: 0.9000
5/5 [==============================] - 0s 627us/step - loss: 0.4370 - accuracy: 0.8958

5/5 [==============================] - 0s 27ms/step - loss: 0.4370 - accuracy: 0.8958 - val_loss: 0.7146 - val_accuracy: 0.6400
Epoch 101/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3194 - accuracy: 1.0000
5/5 [==============================] - 0s 2ms/step - loss: 0.4322 - accuracy: 0.8958

5/5 [==============================] - 0s 31ms/step - loss: 0.4322 - accuracy: 0.8958 - val_loss: 0.7067 - val_accuracy: 0.6400
Epoch 102/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4717 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.4277 - accuracy: 0.8958

5/5 [==============================] - 0s 30ms/step - loss: 0.4277 - accuracy: 0.8958 - val_loss: 0.6990 - val_accuracy: 0.6400
Epoch 103/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4912 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.4235 - accuracy: 0.8854

5/5 [==============================] - 0s 33ms/step - loss: 0.4235 - accuracy: 0.8854 - val_loss: 0.6945 - val_accuracy: 0.6400
Epoch 104/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4160 - accuracy: 0.9500
5/5 [==============================] - 0s 632us/step - loss: 0.4189 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.4189 - accuracy: 0.9062 - val_loss: 0.6884 - val_accuracy: 0.6400
Epoch 105/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4485 - accuracy: 0.9000
5/5 [==============================] - 0s 531us/step - loss: 0.4149 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.4149 - accuracy: 0.9062 - val_loss: 0.6838 - val_accuracy: 0.6400
Epoch 106/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4458 - accuracy: 0.9000
5/5 [==============================] - 0s 794us/step - loss: 0.4107 - accuracy: 0.9167

5/5 [==============================] - 0s 28ms/step - loss: 0.4107 - accuracy: 0.9167 - val_loss: 0.6801 - val_accuracy: 0.6400
Epoch 107/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3733 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.4069 - accuracy: 0.9062

5/5 [==============================] - 0s 30ms/step - loss: 0.4069 - accuracy: 0.9062 - val_loss: 0.6755 - val_accuracy: 0.6400
Epoch 108/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4710 - accuracy: 0.8500
5/5 [==============================] - 0s 1ms/step - loss: 0.4025 - accuracy: 0.9062

5/5 [==============================] - 0s 31ms/step - loss: 0.4025 - accuracy: 0.9062 - val_loss: 0.6685 - val_accuracy: 0.6400
Epoch 109/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3530 - accuracy: 1.0000
5/5 [==============================] - 0s 3ms/step - loss: 0.3987 - accuracy: 0.9062

5/5 [==============================] - 0s 32ms/step - loss: 0.3987 - accuracy: 0.9062 - val_loss: 0.6603 - val_accuracy: 0.6400
Epoch 110/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3945 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.3946 - accuracy: 0.9062

5/5 [==============================] - 0s 29ms/step - loss: 0.3946 - accuracy: 0.9062 - val_loss: 0.6558 - val_accuracy: 0.6400
Epoch 111/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3937 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.3907 - accuracy: 0.9062

5/5 [==============================] - 0s 28ms/step - loss: 0.3907 - accuracy: 0.9062 - val_loss: 0.6508 - val_accuracy: 0.6400
Epoch 112/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3172 - accuracy: 0.9000
5/5 [==============================] - 0s 571us/step - loss: 0.3869 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3869 - accuracy: 0.9062 - val_loss: 0.6435 - val_accuracy: 0.6400
Epoch 113/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4204 - accuracy: 0.9000
5/5 [==============================] - 0s 537us/step - loss: 0.3834 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3834 - accuracy: 0.9062 - val_loss: 0.6407 - val_accuracy: 0.6400
Epoch 114/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3559 - accuracy: 0.9500
5/5 [==============================] - 0s 3ms/step - loss: 0.3807 - accuracy: 0.9062

5/5 [==============================] - 0s 34ms/step - loss: 0.3807 - accuracy: 0.9062 - val_loss: 0.6431 - val_accuracy: 0.6400
Epoch 115/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3245 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.3765 - accuracy: 0.9062

5/5 [==============================] - 0s 31ms/step - loss: 0.3765 - accuracy: 0.9062 - val_loss: 0.6409 - val_accuracy: 0.6400
Epoch 116/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3550 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.3729 - accuracy: 0.9062

5/5 [==============================] - 0s 31ms/step - loss: 0.3729 - accuracy: 0.9062 - val_loss: 0.6369 - val_accuracy: 0.6400
Epoch 117/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3448 - accuracy: 1.0000
5/5 [==============================] - 0s 649us/step - loss: 0.3697 - accuracy: 0.9062

5/5 [==============================] - 0s 28ms/step - loss: 0.3697 - accuracy: 0.9062 - val_loss: 0.6300 - val_accuracy: 0.6400
Epoch 118/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3611 - accuracy: 0.8500
5/5 [==============================] - 0s 649us/step - loss: 0.3664 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3664 - accuracy: 0.9062 - val_loss: 0.6276 - val_accuracy: 0.6400
Epoch 119/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3513 - accuracy: 0.9500
5/5 [==============================] - 0s 582us/step - loss: 0.3634 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3634 - accuracy: 0.9062 - val_loss: 0.6221 - val_accuracy: 0.6400
Epoch 120/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4132 - accuracy: 0.8000
5/5 [==============================] - 0s 1ms/step - loss: 0.3607 - accuracy: 0.9062

5/5 [==============================] - 0s 30ms/step - loss: 0.3607 - accuracy: 0.9062 - val_loss: 0.6159 - val_accuracy: 0.6400
Epoch 121/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3394 - accuracy: 0.9000
5/5 [==============================] - 0s 552us/step - loss: 0.3571 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3571 - accuracy: 0.9062 - val_loss: 0.6110 - val_accuracy: 0.6400
Epoch 122/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2415 - accuracy: 0.9500
5/5 [==============================] - 0s 628us/step - loss: 0.3550 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3550 - accuracy: 0.9062 - val_loss: 0.6072 - val_accuracy: 0.6400
Epoch 123/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3936 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.3514 - accuracy: 0.9062

5/5 [==============================] - 0s 33ms/step - loss: 0.3514 - accuracy: 0.9062 - val_loss: 0.6065 - val_accuracy: 0.6400
Epoch 124/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4507 - accuracy: 0.8500
5/5 [==============================] - 0s 1ms/step - loss: 0.3487 - accuracy: 0.9062

5/5 [==============================] - 0s 29ms/step - loss: 0.3487 - accuracy: 0.9062 - val_loss: 0.6042 - val_accuracy: 0.6800
Epoch 125/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3833 - accuracy: 0.8500
5/5 [==============================] - 0s 605us/step - loss: 0.3459 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3459 - accuracy: 0.9062 - val_loss: 0.6021 - val_accuracy: 0.6800
Epoch 126/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3303 - accuracy: 0.8500
5/5 [==============================] - 0s 1ms/step - loss: 0.3431 - accuracy: 0.9062

5/5 [==============================] - 0s 27ms/step - loss: 0.3431 - accuracy: 0.9062 - val_loss: 0.5988 - val_accuracy: 0.6800
Epoch 127/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2951 - accuracy: 0.9500
5/5 [==============================] - 0s 635us/step - loss: 0.3406 - accuracy: 0.9062

5/5 [==============================] - 0s 27ms/step - loss: 0.3406 - accuracy: 0.9062 - val_loss: 0.5936 - val_accuracy: 0.6800
Epoch 128/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3692 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.3374 - accuracy: 0.9062

5/5 [==============================] - 0s 29ms/step - loss: 0.3374 - accuracy: 0.9062 - val_loss: 0.5912 - val_accuracy: 0.6800
Epoch 129/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3332 - accuracy: 1.0000
5/5 [==============================] - 0s 2ms/step - loss: 0.3351 - accuracy: 0.9062

5/5 [==============================] - 0s 33ms/step - loss: 0.3351 - accuracy: 0.9062 - val_loss: 0.5888 - val_accuracy: 0.6800
Epoch 130/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3401 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.3322 - accuracy: 0.9062

5/5 [==============================] - 0s 30ms/step - loss: 0.3322 - accuracy: 0.9062 - val_loss: 0.5854 - val_accuracy: 0.6800
Epoch 131/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2288 - accuracy: 0.9000
5/5 [==============================] - 0s 784us/step - loss: 0.3294 - accuracy: 0.9062

5/5 [==============================] - 0s 28ms/step - loss: 0.3294 - accuracy: 0.9062 - val_loss: 0.5822 - val_accuracy: 0.6800
Epoch 132/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3352 - accuracy: 0.9500
5/5 [==============================] - 0s 616us/step - loss: 0.3269 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3269 - accuracy: 0.9062 - val_loss: 0.5779 - val_accuracy: 0.6800
Epoch 133/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3572 - accuracy: 0.9000
5/5 [==============================] - 0s 588us/step - loss: 0.3242 - accuracy: 0.9062

5/5 [==============================] - 0s 26ms/step - loss: 0.3242 - accuracy: 0.9062 - val_loss: 0.5736 - val_accuracy: 0.6800
Epoch 134/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3048 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.3220 - accuracy: 0.9167

5/5 [==============================] - 0s 34ms/step - loss: 0.3220 - accuracy: 0.9167 - val_loss: 0.5694 - val_accuracy: 0.7200
Epoch 135/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3022 - accuracy: 0.9500
5/5 [==============================] - 0s 3ms/step - loss: 0.3196 - accuracy: 0.9167

5/5 [==============================] - 0s 33ms/step - loss: 0.3196 - accuracy: 0.9167 - val_loss: 0.5636 - val_accuracy: 0.7200
Epoch 136/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2346 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.3169 - accuracy: 0.9167

5/5 [==============================] - 0s 30ms/step - loss: 0.3169 - accuracy: 0.9167 - val_loss: 0.5574 - val_accuracy: 0.7200
Epoch 137/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3432 - accuracy: 0.9000
5/5 [==============================] - 0s 3ms/step - loss: 0.3146 - accuracy: 0.9167

5/5 [==============================] - 0s 35ms/step - loss: 0.3146 - accuracy: 0.9167 - val_loss: 0.5529 - val_accuracy: 0.7600
Epoch 138/200

1/5 [=====>........................] - ETA: 0s - loss: 0.4000 - accuracy: 0.9000
5/5 [==============================] - 0s 3ms/step - loss: 0.3123 - accuracy: 0.9167

5/5 [==============================] - 0s 32ms/step - loss: 0.3123 - accuracy: 0.9167 - val_loss: 0.5504 - val_accuracy: 0.7600
Epoch 139/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3690 - accuracy: 0.8000
5/5 [==============================] - 0s 606us/step - loss: 0.3106 - accuracy: 0.9167

5/5 [==============================] - 0s 26ms/step - loss: 0.3106 - accuracy: 0.9167 - val_loss: 0.5493 - val_accuracy: 0.7600
Epoch 140/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2240 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.3079 - accuracy: 0.9167

5/5 [==============================] - 0s 32ms/step - loss: 0.3079 - accuracy: 0.9167 - val_loss: 0.5444 - val_accuracy: 0.7600
Epoch 141/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2609 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.3058 - accuracy: 0.9271

5/5 [==============================] - 0s 31ms/step - loss: 0.3058 - accuracy: 0.9271 - val_loss: 0.5440 - val_accuracy: 0.7600
Epoch 142/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2301 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.3039 - accuracy: 0.9167

5/5 [==============================] - 0s 31ms/step - loss: 0.3039 - accuracy: 0.9167 - val_loss: 0.5399 - val_accuracy: 0.7600
Epoch 143/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2970 - accuracy: 1.0000
5/5 [==============================] - 0s 924us/step - loss: 0.3015 - accuracy: 0.9271

5/5 [==============================] - 0s 30ms/step - loss: 0.3015 - accuracy: 0.9271 - val_loss: 0.5362 - val_accuracy: 0.8000
Epoch 144/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2397 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.2992 - accuracy: 0.9271

5/5 [==============================] - 0s 31ms/step - loss: 0.2992 - accuracy: 0.9271 - val_loss: 0.5318 - val_accuracy: 0.8000
Epoch 145/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2133 - accuracy: 0.9500
5/5 [==============================] - 0s 649us/step - loss: 0.2973 - accuracy: 0.9375

5/5 [==============================] - 0s 26ms/step - loss: 0.2973 - accuracy: 0.9375 - val_loss: 0.5313 - val_accuracy: 0.8000
Epoch 146/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3304 - accuracy: 0.8500
5/5 [==============================] - 0s 610us/step - loss: 0.2948 - accuracy: 0.9375

5/5 [==============================] - 0s 28ms/step - loss: 0.2948 - accuracy: 0.9375 - val_loss: 0.5286 - val_accuracy: 0.8000
Epoch 147/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1925 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2933 - accuracy: 0.9375

5/5 [==============================] - 0s 30ms/step - loss: 0.2933 - accuracy: 0.9375 - val_loss: 0.5279 - val_accuracy: 0.8000
Epoch 148/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2841 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.2906 - accuracy: 0.9375

5/5 [==============================] - 0s 62ms/step - loss: 0.2906 - accuracy: 0.9375 - val_loss: 0.5264 - val_accuracy: 0.8000
Epoch 149/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1974 - accuracy: 1.0000
5/5 [==============================] - 0s 2ms/step - loss: 0.2891 - accuracy: 0.9375

5/5 [==============================] - 0s 31ms/step - loss: 0.2891 - accuracy: 0.9375 - val_loss: 0.5256 - val_accuracy: 0.8000
Epoch 150/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1827 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2866 - accuracy: 0.9375

5/5 [==============================] - 0s 30ms/step - loss: 0.2866 - accuracy: 0.9375 - val_loss: 0.5201 - val_accuracy: 0.8000
Epoch 151/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1956 - accuracy: 1.0000
5/5 [==============================] - 0s 634us/step - loss: 0.2849 - accuracy: 0.9375

5/5 [==============================] - 0s 26ms/step - loss: 0.2849 - accuracy: 0.9375 - val_loss: 0.5148 - val_accuracy: 0.8000
Epoch 152/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2623 - accuracy: 0.9500
5/5 [==============================] - 0s 592us/step - loss: 0.2828 - accuracy: 0.9375

5/5 [==============================] - 0s 26ms/step - loss: 0.2828 - accuracy: 0.9375 - val_loss: 0.5116 - val_accuracy: 0.8000
Epoch 153/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3153 - accuracy: 0.9000
5/5 [==============================] - 0s 994us/step - loss: 0.2807 - accuracy: 0.9375

5/5 [==============================] - 0s 30ms/step - loss: 0.2807 - accuracy: 0.9375 - val_loss: 0.5087 - val_accuracy: 0.8000
Epoch 154/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1895 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2786 - accuracy: 0.9375

5/5 [==============================] - 0s 32ms/step - loss: 0.2786 - accuracy: 0.9375 - val_loss: 0.5035 - val_accuracy: 0.8000
Epoch 155/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2475 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2764 - accuracy: 0.9479

5/5 [==============================] - 0s 33ms/step - loss: 0.2764 - accuracy: 0.9479 - val_loss: 0.4989 - val_accuracy: 0.8000
Epoch 156/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1431 - accuracy: 1.0000
5/5 [==============================] - 0s 2ms/step - loss: 0.2747 - accuracy: 0.9479

5/5 [==============================] - 0s 34ms/step - loss: 0.2747 - accuracy: 0.9479 - val_loss: 0.4969 - val_accuracy: 0.8000
Epoch 157/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2316 - accuracy: 0.9500
5/5 [==============================] - 0s 625us/step - loss: 0.2724 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2724 - accuracy: 0.9479 - val_loss: 0.4940 - val_accuracy: 0.8000
Epoch 158/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3008 - accuracy: 0.9500
5/5 [==============================] - 0s 611us/step - loss: 0.2707 - accuracy: 0.9479

5/5 [==============================] - 0s 27ms/step - loss: 0.2707 - accuracy: 0.9479 - val_loss: 0.4938 - val_accuracy: 0.8000
Epoch 159/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2216 - accuracy: 0.9500
5/5 [==============================] - 0s 737us/step - loss: 0.2689 - accuracy: 0.9375

5/5 [==============================] - 0s 28ms/step - loss: 0.2689 - accuracy: 0.9375 - val_loss: 0.4888 - val_accuracy: 0.8400
Epoch 160/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3343 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2668 - accuracy: 0.9479

5/5 [==============================] - 0s 31ms/step - loss: 0.2668 - accuracy: 0.9479 - val_loss: 0.4877 - val_accuracy: 0.8400
Epoch 161/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3573 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.2650 - accuracy: 0.9479

5/5 [==============================] - 0s 30ms/step - loss: 0.2650 - accuracy: 0.9479 - val_loss: 0.4859 - val_accuracy: 0.8400
Epoch 162/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2741 - accuracy: 1.0000
5/5 [==============================] - 0s 849us/step - loss: 0.2634 - accuracy: 0.9375

5/5 [==============================] - 0s 28ms/step - loss: 0.2634 - accuracy: 0.9375 - val_loss: 0.4803 - val_accuracy: 0.8400
Epoch 163/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2887 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2614 - accuracy: 0.9479

5/5 [==============================] - 0s 32ms/step - loss: 0.2614 - accuracy: 0.9479 - val_loss: 0.4774 - val_accuracy: 0.8400
Epoch 164/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1815 - accuracy: 1.0000
5/5 [==============================] - 0s 585us/step - loss: 0.2593 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2593 - accuracy: 0.9479 - val_loss: 0.4731 - val_accuracy: 0.8400
Epoch 165/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3202 - accuracy: 0.9000
5/5 [==============================] - 0s 659us/step - loss: 0.2576 - accuracy: 0.9479

5/5 [==============================] - 0s 27ms/step - loss: 0.2576 - accuracy: 0.9479 - val_loss: 0.4721 - val_accuracy: 0.8400
Epoch 166/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3690 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.2554 - accuracy: 0.9479

5/5 [==============================] - 0s 32ms/step - loss: 0.2554 - accuracy: 0.9479 - val_loss: 0.4685 - val_accuracy: 0.8800
Epoch 167/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2891 - accuracy: 0.9500
5/5 [==============================] - 0s 535us/step - loss: 0.2540 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2540 - accuracy: 0.9479 - val_loss: 0.4702 - val_accuracy: 0.8800
Epoch 168/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2278 - accuracy: 1.0000
5/5 [==============================] - 0s 624us/step - loss: 0.2520 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2520 - accuracy: 0.9479 - val_loss: 0.4678 - val_accuracy: 0.8800
Epoch 169/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2974 - accuracy: 0.8500
5/5 [==============================] - 0s 3ms/step - loss: 0.2502 - accuracy: 0.9479

5/5 [==============================] - 0s 32ms/step - loss: 0.2502 - accuracy: 0.9479 - val_loss: 0.4630 - val_accuracy: 0.8800
Epoch 170/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3433 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.2481 - accuracy: 0.9479

5/5 [==============================] - 0s 31ms/step - loss: 0.2481 - accuracy: 0.9479 - val_loss: 0.4595 - val_accuracy: 0.8800
Epoch 171/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2853 - accuracy: 0.9500
5/5 [==============================] - 0s 532us/step - loss: 0.2462 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2462 - accuracy: 0.9479 - val_loss: 0.4594 - val_accuracy: 0.8800
Epoch 172/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2316 - accuracy: 1.0000
5/5 [==============================] - 0s 672us/step - loss: 0.2443 - accuracy: 0.9479

5/5 [==============================] - 0s 26ms/step - loss: 0.2443 - accuracy: 0.9479 - val_loss: 0.4565 - val_accuracy: 0.8800
Epoch 173/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2700 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2432 - accuracy: 0.9583

5/5 [==============================] - 0s 31ms/step - loss: 0.2432 - accuracy: 0.9583 - val_loss: 0.4547 - val_accuracy: 0.8800
Epoch 174/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3305 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.2408 - accuracy: 0.9479

5/5 [==============================] - 0s 30ms/step - loss: 0.2408 - accuracy: 0.9479 - val_loss: 0.4517 - val_accuracy: 0.8800
Epoch 175/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2469 - accuracy: 0.9000
5/5 [==============================] - 0s 2ms/step - loss: 0.2393 - accuracy: 0.9583

5/5 [==============================] - 0s 32ms/step - loss: 0.2393 - accuracy: 0.9583 - val_loss: 0.4505 - val_accuracy: 0.8800
Epoch 176/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2440 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2372 - accuracy: 0.9583

5/5 [==============================] - 0s 34ms/step - loss: 0.2372 - accuracy: 0.9583 - val_loss: 0.4497 - val_accuracy: 0.8800
Epoch 177/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1796 - accuracy: 1.0000
5/5 [==============================] - 0s 669us/step - loss: 0.2361 - accuracy: 0.9583

5/5 [==============================] - 0s 26ms/step - loss: 0.2361 - accuracy: 0.9583 - val_loss: 0.4492 - val_accuracy: 0.8800
Epoch 178/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2099 - accuracy: 0.9500
5/5 [==============================] - 0s 553us/step - loss: 0.2341 - accuracy: 0.9583

5/5 [==============================] - 0s 27ms/step - loss: 0.2341 - accuracy: 0.9583 - val_loss: 0.4442 - val_accuracy: 0.8800
Epoch 179/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2318 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.2327 - accuracy: 0.9583

5/5 [==============================] - 0s 31ms/step - loss: 0.2327 - accuracy: 0.9583 - val_loss: 0.4436 - val_accuracy: 0.8800
Epoch 180/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2123 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2312 - accuracy: 0.9583

5/5 [==============================] - 0s 30ms/step - loss: 0.2312 - accuracy: 0.9583 - val_loss: 0.4430 - val_accuracy: 0.8800
Epoch 181/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2616 - accuracy: 0.9500
5/5 [==============================] - 0s 557us/step - loss: 0.2298 - accuracy: 0.9583

5/5 [==============================] - 0s 28ms/step - loss: 0.2298 - accuracy: 0.9583 - val_loss: 0.4432 - val_accuracy: 0.8800
Epoch 182/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2030 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.2279 - accuracy: 0.9583

5/5 [==============================] - 0s 30ms/step - loss: 0.2279 - accuracy: 0.9583 - val_loss: 0.4407 - val_accuracy: 0.8800
Epoch 183/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3550 - accuracy: 0.8500
5/5 [==============================] - 0s 2ms/step - loss: 0.2272 - accuracy: 0.9583

5/5 [==============================] - 0s 31ms/step - loss: 0.2272 - accuracy: 0.9583 - val_loss: 0.4387 - val_accuracy: 0.8800
Epoch 184/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2117 - accuracy: 1.0000
5/5 [==============================] - 0s 544us/step - loss: 0.2250 - accuracy: 0.9583

5/5 [==============================] - 0s 26ms/step - loss: 0.2250 - accuracy: 0.9583 - val_loss: 0.4361 - val_accuracy: 0.8800
Epoch 185/200

1/5 [=====>........................] - ETA: 0s - loss: 0.3096 - accuracy: 0.9000
5/5 [==============================] - 0s 600us/step - loss: 0.2237 - accuracy: 0.9583

5/5 [==============================] - 0s 30ms/step - loss: 0.2237 - accuracy: 0.9583 - val_loss: 0.4319 - val_accuracy: 0.8800
Epoch 186/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2136 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2222 - accuracy: 0.9583

5/5 [==============================] - 0s 32ms/step - loss: 0.2222 - accuracy: 0.9583 - val_loss: 0.4301 - val_accuracy: 0.8800
Epoch 187/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2844 - accuracy: 0.9500
5/5 [==============================] - 0s 1ms/step - loss: 0.2206 - accuracy: 0.9583

5/5 [==============================] - 0s 31ms/step - loss: 0.2206 - accuracy: 0.9583 - val_loss: 0.4298 - val_accuracy: 0.8800
Epoch 188/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2844 - accuracy: 0.9000
5/5 [==============================] - 0s 1ms/step - loss: 0.2193 - accuracy: 0.9583

5/5 [==============================] - 0s 31ms/step - loss: 0.2193 - accuracy: 0.9583 - val_loss: 0.4287 - val_accuracy: 0.8800
Epoch 189/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1490 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2177 - accuracy: 0.9583

5/5 [==============================] - 0s 30ms/step - loss: 0.2177 - accuracy: 0.9583 - val_loss: 0.4241 - val_accuracy: 0.8800
Epoch 190/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1908 - accuracy: 1.0000
5/5 [==============================] - 0s 793us/step - loss: 0.2167 - accuracy: 0.9583

5/5 [==============================] - 0s 29ms/step - loss: 0.2167 - accuracy: 0.9583 - val_loss: 0.4221 - val_accuracy: 0.8800
Epoch 191/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1632 - accuracy: 1.0000
5/5 [==============================] - 0s 647us/step - loss: 0.2147 - accuracy: 0.9583

5/5 [==============================] - 0s 26ms/step - loss: 0.2147 - accuracy: 0.9583 - val_loss: 0.4198 - val_accuracy: 0.8800
Epoch 192/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1862 - accuracy: 1.0000
5/5 [==============================] - 0s 694us/step - loss: 0.2135 - accuracy: 0.9583

5/5 [==============================] - 0s 27ms/step - loss: 0.2135 - accuracy: 0.9583 - val_loss: 0.4190 - val_accuracy: 0.8800
Epoch 193/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1209 - accuracy: 1.0000
5/5 [==============================] - 0s 1ms/step - loss: 0.2121 - accuracy: 0.9583

5/5 [==============================] - 0s 30ms/step - loss: 0.2121 - accuracy: 0.9583 - val_loss: 0.4193 - val_accuracy: 0.8800
Epoch 194/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1858 - accuracy: 1.0000
5/5 [==============================] - 0s 2ms/step - loss: 0.2103 - accuracy: 0.9583

5/5 [==============================] - 0s 33ms/step - loss: 0.2103 - accuracy: 0.9583 - val_loss: 0.4149 - val_accuracy: 0.8800
Epoch 195/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2436 - accuracy: 0.9500
5/5 [==============================] - 0s 898us/step - loss: 0.2091 - accuracy: 0.9583

5/5 [==============================] - 0s 29ms/step - loss: 0.2091 - accuracy: 0.9583 - val_loss: 0.4117 - val_accuracy: 0.8800
Epoch 196/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1953 - accuracy: 0.9500
5/5 [==============================] - 0s 3ms/step - loss: 0.2080 - accuracy: 0.9583

5/5 [==============================] - 0s 33ms/step - loss: 0.2080 - accuracy: 0.9583 - val_loss: 0.4088 - val_accuracy: 0.8800
Epoch 197/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1973 - accuracy: 1.0000
5/5 [==============================] - 0s 617us/step - loss: 0.2064 - accuracy: 0.9583

5/5 [==============================] - 0s 26ms/step - loss: 0.2064 - accuracy: 0.9583 - val_loss: 0.4097 - val_accuracy: 0.8800
Epoch 198/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1385 - accuracy: 1.0000
5/5 [==============================] - 0s 901us/step - loss: 0.2050 - accuracy: 0.9583

5/5 [==============================] - 0s 29ms/step - loss: 0.2050 - accuracy: 0.9583 - val_loss: 0.4058 - val_accuracy: 0.8800
Epoch 199/200

1/5 [=====>........................] - ETA: 0s - loss: 0.1962 - accuracy: 0.9500
5/5 [==============================] - 0s 2ms/step - loss: 0.2034 - accuracy: 0.9583

5/5 [==============================] - 0s 32ms/step - loss: 0.2034 - accuracy: 0.9583 - val_loss: 0.4059 - val_accuracy: 0.8800
Epoch 200/200

1/5 [=====>........................] - ETA: 0s - loss: 0.2389 - accuracy: 0.9500
5/5 [==============================] - 0s 3ms/step - loss: 0.2023 - accuracy: 0.9583

5/5 [==============================] - 0s 32ms/step - loss: 0.2023 - accuracy: 0.9583 - val_loss: 0.4052 - val_accuracy: 0.8800
plot(history) +
  ggtitle("Training a neural network based classifier on the iris data set") +
  theme_bw()

Evaluate Network Performance

The final performance can be obtained like so.

perf <- model %>% evaluate(x_test, y_test)

1/1 [==============================] - 0s 10us/step - loss: 0.2131 - accuracy: 0.9310

1/1 [==============================] - 0s 190us/step - loss: 0.2131 - accuracy: 0.9310
print(perf)
     loss  accuracy 
0.2130527 0.9310345 

For the next plot the predicted and true values need to be in a vector. Note that the true values need to be unlisted before putting them into a numeric vector.

classes <- iris %>% pull(Species) %>% unique()
y_pred  <- model %>% predict_classes(x_test)
y_true  <- test %>% select(class_label) %>% unlist() %>% as.numeric()

tibble(y_true = classes[y_true + 1], y_pred = classes[y_pred + 1],
       Correct = ifelse(y_true == y_pred, "Yes", "No") %>% factor) %>% 
  ggplot(aes(x = y_true, y = y_pred, colour = Correct)) +
  geom_jitter() +
  theme_bw() +
  ggtitle(label = "Classification Performance of Artificial Neural Network",
          subtitle = str_c("Accuracy = ",round(perf[2],3)*100,"%")) +
  xlab(label = "True iris class") +
  ylab(label = "Predicted iris class")

library(gmodels)

CrossTable(y_pred, y_true,
           prop.chisq = FALSE, prop.t = FALSE, prop.r = FALSE,
           dnn = c('predicted', 'actual'))

 
   Cell Contents
|-------------------------|
|                       N |
|           N / Col Total |
|-------------------------|

 
Total Observations in Table:  29 

 
             | actual 
   predicted |         0 |         1 |         2 | Row Total | 
-------------|-----------|-----------|-----------|-----------|
           0 |        11 |         0 |         0 |        11 | 
             |     1.000 |     0.000 |     0.000 |           | 
-------------|-----------|-----------|-----------|-----------|
           1 |         0 |        10 |         0 |        10 | 
             |     0.000 |     0.833 |     0.000 |           | 
-------------|-----------|-----------|-----------|-----------|
           2 |         0 |         2 |         6 |         8 | 
             |     0.000 |     0.167 |     1.000 |           | 
-------------|-----------|-----------|-----------|-----------|
Column Total |        11 |        12 |         6 |        29 | 
             |     0.379 |     0.414 |     0.207 |           | 
-------------|-----------|-----------|-----------|-----------|

 

Conclusion

I hope this illustrated just how easy it is to get started building artificial neural network using Keras and TensorFlow in R. With relative ease, we created a 3-class predictor with an accuracy of 100%. This was a basic minimal example. The network can be expanded to create Deep Learning networks and also the entire TensorFlow API is available.

Enjoy and Happy Learning!

Leon

Thanks again Leon, this was awesome!!!

---
title: "Building a simple neural network using Keras and Tensorflow - Updated"
output:
  word_document: default
  html_notebook: default
  pdf_document: default
  html_document:
    df_print: paged
---

**Update:** The original code has been updated to use the *tidymodels* init_split() function, rather than using the indicies method which originally used setdiff, which now may have a conflict between base R and the tidyverse.

Thank you

A big thank you to Leon Jessen for posting his code on github.

[Building a simple neural network using Keras and Tensorflow](https://github.com/leonjessen/keras_tensorflow_on_iris/blob/master/README.md)

I have forked his project on github and put his code into an R Notebook so we can run it in class.

### Motivation

The following is a minimal example for building your first simple artificial neural network using Keras and TensorFlow for R.

[TensorFlow for R by Rstudio lives here](https://tensorflow.rstudio.com/keras/).

### Gettings started - Install Keras and TensorFlow for R

You can install the Keras for R package from CRAN as follows:

```{r eval=FALSE}
# install.packages("keras")
```

TensorFlow is the default backend engine. TensorFlow and Keras can be installed as follows:

```{r eval=FALSE}
# library(keras)
# install_keras()
```

Naturally, we will also need **Tidyverse**.

```{r eval=FALSE}
# Install from CRAN
# install.packages("tidyverse")

# Or the development version from GitHub
# install.packages("devtools")
# devtools::install_github("hadley/tidyverse")
```

Once installed, we simply load the libraries.

```{r}
library("keras")
suppressMessages(library("tidyverse"))
```

### Artificial Neural Network Using the Iris Data Set

Right, let's get to it!

### Data

The famous (Fisher's or Anderson's) *iris* data set contains a total of 150 observations of 4 input features *Sepal.Length*, *Sepal.Width*, *Petal.Length* and *Petal.Width* and 3 output classes *setosa* *versicolor* and *virginica*, with 50 observations in each class. The distributions of the feature values looks like so:

```{r}
iris_tib <- as_tibble(iris)
iris_tib
```

```{r}
iris_tib %>% pivot_longer(names_to = "feature", values_to = "value", -Species) %>%
  ggplot(aes(x = feature, y = value, fill = Species)) +
  geom_violin(alpha = 0.5, scale = "width") +
  theme_bw()
```

Our aim is to connect the 4 input features to the correct output class using an artificial neural network. For this task, we have chosen the following simple architecture with one input layer with 4 neurons (one for each feature), one hidden layer with 4 neurons and one output layer with 3 neurons (one for each class), all fully connected.

![architecture_visualisation.png](/home/esuess/Documents/Stat654/iris_nn/img/architecture_visualisation.png)

Our artificial neural network will have a total of 35 parameters: 4 for each input neuron connected to the hidden layer, plus an additional 4 for the associated first bias neuron and 3 for each of the hidden neurons connected to the output layer, plus an additional 3 for the associated second bias neuron, i.e. $4 \times 4 + 4 + 4 \times 3 + 3=35$

### Prepare data

We start with slightly wrangling the iris data set by renaming and scaling the features and converting character labels to numeric.

```{r}
set.seed(265509)
nn_dat <- iris_tib %>%
  mutate(sepal_length = scale(Sepal.Length),
         sepal_width  = scale(Sepal.Width),
         petal_length = scale(Petal.Length),
         petal_width  = scale(Petal.Width),          
         class_label  = as.numeric(Species) - 1) %>% 
    select(sepal_length, sepal_width, petal_length, petal_width, class_label)

nn_dat %>% head()
```

Then, we create indices for splitting the iris data into a training and a test data set. We set aside 20% of the data for testing.

```{r}
library(tidymodels)

set.seed(364)
n <- nrow(nn_dat)
n

iris_parts <- nn_dat %>%
  initial_split(prop = 0.8)

train <- iris_parts %>%
  training()

test <- iris_parts %>%
  testing()

list(train, test) %>%
  map_int(nrow)
```

```{r}

n_total_samples <- nrow(nn_dat)

n_train_samples <- nrow(train)

n_test_samples <- nrow(test)

```

### Create training and test data

**Note** that the functions in the keras package are expecting the data to be in a matrix object and not a tibble.  So as.matrix is added at the end of each line.

```{r}
x_train <- train %>% select(-class_label) %>% as.matrix()
y_train <- train %>% select(class_label) %>% as.matrix() %>% to_categorical()

x_test <- test %>% select(-class_label) %>% as.matrix()
y_test <- test %>% select(class_label) %>% as.matrix() %>% to_categorical() 

dim(y_train)
dim(y_test)
```

### Set Architecture

With the data in place, we now set the architecture of our neural network.

```{r}
model <- keras_model_sequential()
model %>% 
  layer_dense(units = 4, activation = 'relu', input_shape = 4) %>% 
  layer_dense(units = 3, activation = 'softmax')
model %>% summary
```

Next, the architecture set in the model needs to be compiled.

```{r}
model %>% compile(
  loss      = 'categorical_crossentropy',
  optimizer = optimizer_rmsprop(),
  metrics   = c('accuracy')
)
```

### Train the Artificial Neural Network

Lastly we fit the model and save the training progress in the *history* object.

**Try** changing the *validation_split* from 0 to 0.2 to see the *validation_loss*.


```{r}
history <- model %>% fit(
  x = x_train, y = y_train,
  epochs = 200,
  batch_size = 20,
  validation_split = 0.2
)

plot(history) +
  ggtitle("Training a neural network based classifier on the iris data set") +
  theme_bw()
```

### Evaluate Network Performance

The final performance can be obtained like so.

```{r}
perf <- model %>% evaluate(x_test, y_test)
print(perf)
```

For the next plot the predicted and true values need to be in a vector.  Note that the true values need to be unlisted before putting them into a numeric vector. 

```{r}
classes <- iris %>% pull(Species) %>% unique()
y_pred  <- model %>% predict_classes(x_test)
y_true  <- test %>% select(class_label) %>% unlist() %>% as.numeric()

tibble(y_true = classes[y_true + 1], y_pred = classes[y_pred + 1],
       Correct = ifelse(y_true == y_pred, "Yes", "No") %>% factor) %>% 
  ggplot(aes(x = y_true, y = y_pred, colour = Correct)) +
  geom_jitter() +
  theme_bw() +
  ggtitle(label = "Classification Performance of Artificial Neural Network",
          subtitle = str_c("Accuracy = ",round(perf[2],3)*100,"%")) +
  xlab(label = "True iris class") +
  ylab(label = "Predicted iris class")
```


```{r}
library(gmodels)

CrossTable(y_pred, y_true,
           prop.chisq = FALSE, prop.t = FALSE, prop.r = FALSE,
           dnn = c('predicted', 'actual'))

```


### Conclusion

I hope this illustrated just how easy it is to get started building artificial neural network using Keras and TensorFlow in R. With relative ease, we created a 3-class predictor with an accuracy of 100%. This was a basic minimal example. The network can be expanded to create Deep Learning networks and also the entire TensorFlow API is available.

Enjoy and Happy Learning!

Leon

**Thanks again Leon, this was awesome!!!**