Producing photos with Keras and TensorFlow keen execution

The latest announcement of TensorFlow 2.0 names keen execution because the primary central function of the brand new main model. What does this imply for R customers?
As demonstrated in our latest publish on neural machine translation, you need to use keen execution from R now already, together with Keras customized fashions and the datasets API. It’s good to know you can use it – however why do you have to? And during which circumstances?

On this and some upcoming posts, we wish to present how keen execution could make growing fashions loads simpler. The diploma of simplication will rely upon the duty – and simply how a lot simpler you’ll discover the brand new manner may also rely in your expertise utilizing the practical API to mannequin extra complicated relationships.
Even in the event you assume that GANs, encoder-decoder architectures, or neural model switch didn’t pose any issues earlier than the arrival of keen execution, you would possibly discover that the choice is a greater match to how we people mentally image issues.

For this publish, we’re porting code from a latest Google Colaboratory pocket book implementing the DCGAN structure.(Radford, Metz, and Chintala 2015)
No prior data of GANs is required – we’ll hold this publish sensible (no maths) and give attention to easy methods to obtain your aim, mapping a easy and vivid idea into an astonishingly small variety of traces of code.

As within the publish on machine translation with consideration, we first should cowl some conditions.
By the best way, no want to repeat out the code snippets – you’ll discover the entire code in eager_dcgan.R).

Conditions

The code on this publish is determined by the most recent CRAN variations of a number of of the TensorFlow R packages. You may set up these packages as follows:

set up.packages(c("tensorflow", "keras", "tfdatasets"))

library(tensorflow)
install_tensorflow()

We’ll additionally use the tfdatasets bundle for our enter pipeline. So we find yourself with the next preamble to set issues up:

That’s it. Let’s get began.

So what’s a GAN?

GAN stands for Generative Adversarial Community(Goodfellow et al. 2014). It’s a setup of two brokers, the generator and the discriminator, that act in opposition to one another (thus, adversarial). It’s generative as a result of the aim is to generate output (versus, say, classification or regression).

In human studying, suggestions – direct or oblique – performs a central position. Say we wished to forge a banknote (so long as these nonetheless exist). Assuming we will get away with unsuccessful trials, we might get higher and higher at forgery over time. Optimizing our method, we might find yourself wealthy.
This idea of optimizing from suggestions is embodied within the first of the 2 brokers, the generator. It will get its suggestions from the discriminator, in an upside-down manner: If it may idiot the discriminator, making it consider that the banknote was actual, all is ok; if the discriminator notices the pretend, it has to do issues otherwise. For a neural community, meaning it has to replace its weights.

How does the discriminator know what’s actual and what’s pretend? It too must be skilled, on actual banknotes (or regardless of the sort of objects concerned) and the pretend ones produced by the generator. So the entire setup is 2 brokers competing, one striving to generate realistic-looking pretend objects, and the opposite, to disavow the deception. The aim of coaching is to have each evolve and get higher, in flip inflicting the opposite to get higher, too.

On this system, there is no such thing as a goal minimal to the loss operate: We wish each elements to be taught and getter higher “in lockstep,” as a substitute of 1 successful out over the opposite. This makes optimization tough.
In observe due to this fact, tuning a GAN can appear extra like alchemy than like science, and it usually is sensible to lean on practices and “tips” reported by others.

On this instance, identical to within the Google pocket book we’re porting, the aim is to generate MNIST digits. Whereas that won’t sound like essentially the most thrilling process one may think about, it lets us give attention to the mechanics, and permits us to maintain computation and reminiscence necessities (comparatively) low.

Let’s load the information (coaching set wanted solely) after which, take a look at the primary actor in our drama, the generator.

Coaching knowledge

mnist <- dataset_mnist()
c(train_images, train_labels) %<-% mnist$practice

train_images <- train_images %>% 
  k_expand_dims() %>%
  k_cast(dtype = "float32")

# normalize photos to [-1, 1] as a result of the generator makes use of tanh activation
train_images <- (train_images - 127.5) / 127.5

Our full coaching set shall be streamed as soon as per epoch:

buffer_size <- 60000
batch_size <- 256
batches_per_epoch <- (buffer_size / batch_size) %>% spherical()

train_dataset <- tensor_slices_dataset(train_images) %>%
  dataset_shuffle(buffer_size) %>%
  dataset_batch(batch_size)

This enter shall be fed to the discriminator solely.

Generator

Each generator and discriminator are Keras customized fashions.
In distinction to customized layers, customized fashions mean you can assemble fashions as impartial items, full with customized ahead cross logic, backprop and optimization. The model-generating operate defines the layers the mannequin (self) needs assigned, and returns the operate that implements the ahead cross.

As we’ll quickly see, the generator will get handed vectors of random noise for enter. This vector is remodeled to 3d (peak, width, channels) after which, successively upsampled to the required output dimension of (28,28,3).

generator <-
  operate(identify = NULL) {
    keras_model_custom(identify = identify, operate(self) {
      
      self$fc1 <- layer_dense(items = 7 * 7 * 64, use_bias = FALSE)
      self$batchnorm1 <- layer_batch_normalization()
      self$leaky_relu1 <- layer_activation_leaky_relu()
      self$conv1 <-
        layer_conv_2d_transpose(
          filters = 64,
          kernel_size = c(5, 5),
          strides = c(1, 1),
          padding = "similar",
          use_bias = FALSE
        )
      self$batchnorm2 <- layer_batch_normalization()
      self$leaky_relu2 <- layer_activation_leaky_relu()
      self$conv2 <-
        layer_conv_2d_transpose(
          filters = 32,
          kernel_size = c(5, 5),
          strides = c(2, 2),
          padding = "similar",
          use_bias = FALSE
        )
      self$batchnorm3 <- layer_batch_normalization()
      self$leaky_relu3 <- layer_activation_leaky_relu()
      self$conv3 <-
        layer_conv_2d_transpose(
          filters = 1,
          kernel_size = c(5, 5),
          strides = c(2, 2),
          padding = "similar",
          use_bias = FALSE,
          activation = "tanh"
        )
      
      operate(inputs, masks = NULL, coaching = TRUE) {
        self$fc1(inputs) %>%
          self$batchnorm1(coaching = coaching) %>%
          self$leaky_relu1() %>%
          k_reshape(form = c(-1, 7, 7, 64)) %>%
          self$conv1() %>%
          self$batchnorm2(coaching = coaching) %>%
          self$leaky_relu2() %>%
          self$conv2() %>%
          self$batchnorm3(coaching = coaching) %>%
          self$leaky_relu3() %>%
          self$conv3()
      }
    })
  }

Discriminator

The discriminator is only a fairly regular convolutional community outputting a rating. Right here, utilization of “rating” as a substitute of “likelihood” is on objective: If you happen to take a look at the final layer, it’s absolutely related, of dimension 1 however missing the same old sigmoid activation. It is because in contrast to Keras’ loss_binary_crossentropy, the loss operate we’ll be utilizing right here – tf$losses$sigmoid_cross_entropy – works with the uncooked logits, not the outputs of the sigmoid.

discriminator <-
  operate(identify = NULL) {
    keras_model_custom(identify = identify, operate(self) {
      
      self$conv1 <- layer_conv_2d(
        filters = 64,
        kernel_size = c(5, 5),
        strides = c(2, 2),
        padding = "similar"
      )
      self$leaky_relu1 <- layer_activation_leaky_relu()
      self$dropout <- layer_dropout(charge = 0.3)
      self$conv2 <-
        layer_conv_2d(
          filters = 128,
          kernel_size = c(5, 5),
          strides = c(2, 2),
          padding = "similar"
        )
      self$leaky_relu2 <- layer_activation_leaky_relu()
      self$flatten <- layer_flatten()
      self$fc1 <- layer_dense(items = 1)
      
      operate(inputs, masks = NULL, coaching = TRUE) {
        inputs %>% self$conv1() %>%
          self$leaky_relu1() %>%
          self$dropout(coaching = coaching) %>%
          self$conv2() %>%
          self$leaky_relu2() %>%
          self$flatten() %>%
          self$fc1()
      }
    })
  }

Setting the scene

Earlier than we will begin coaching, we have to create the same old elements of a deep studying setup: the mannequin (or fashions, on this case), the loss operate(s), and the optimizer(s).

Mannequin creation is only a operate name, with just a little further on prime:

generator <- generator()
discriminator <- discriminator()

# https://www.tensorflow.org/api_docs/python/tf/contrib/keen/defun
generator$name = tf$contrib$keen$defun(generator$name)
discriminator$name = tf$contrib$keen$defun(discriminator$name)

defun compiles an R operate (as soon as per totally different mixture of argument shapes and non-tensor objects values)) right into a TensorFlow graph, and is used to hurry up computations. This comes with uncomfortable side effects and presumably surprising conduct – please seek the advice of the documentation for the main points. Right here, we had been primarily curious in how a lot of a speedup we would discover when utilizing this from R – in our instance, it resulted in a speedup of 130%.

On to the losses. Discriminator loss consists of two components: Does it accurately determine actual photos as actual, and does it accurately spot pretend photos as pretend.
Right here real_output and generated_output comprise the logits returned from the discriminator – that’s, its judgment of whether or not the respective photos are pretend or actual.

discriminator_loss <- operate(real_output, generated_output) {
  real_loss <- tf$losses$sigmoid_cross_entropy(
    multi_class_labels = k_ones_like(real_output),
    logits = real_output)
  generated_loss <- tf$losses$sigmoid_cross_entropy(
    multi_class_labels = k_zeros_like(generated_output),
    logits = generated_output)
  real_loss + generated_loss
}

Generator loss is determined by how the discriminator judged its creations: It could hope for all of them to be seen as actual.

generator_loss <- operate(generated_output) {
  tf$losses$sigmoid_cross_entropy(
    tf$ones_like(generated_output),
    generated_output)
}

Now we nonetheless have to outline optimizers, one for every mannequin.

discriminator_optimizer <- tf$practice$AdamOptimizer(1e-4)
generator_optimizer <- tf$practice$AdamOptimizer(1e-4)

Coaching loop

There are two fashions, two loss capabilities and two optimizers, however there is only one coaching loop, as each fashions rely upon one another.
The coaching loop shall be over MNIST photos streamed in batches, however we nonetheless want enter to the generator – a random vector of dimension 100, on this case.

Let’s take the coaching loop step-by-step.
There shall be an outer and an internal loop, one over epochs and one over batches.
Firstly of every epoch, we create a recent iterator over the dataset:

for (epoch in seq_len(num_epochs)) {
  begin <- Sys.time()
  total_loss_gen <- 0
  total_loss_disc <- 0
  iter <- make_iterator_one_shot(train_dataset)

until_out_of_range({
  batch <- iterator_get_next(iter)
  noise <- k_random_normal(c(batch_size, noise_dim))
  with(tf$GradientTape() %as% gen_tape, { with(tf$GradientTape() %as% disc_tape, {
    generated_images <- generator(noise)
    disc_real_output <- discriminator(batch, coaching = TRUE)
    disc_generated_output <-
       discriminator(generated_images, coaching = TRUE)
    gen_loss <- generator_loss(disc_generated_output)
    disc_loss <- discriminator_loss(disc_real_output, disc_generated_output)
  }) })

gradients_of_generator <-
  gen_tape$gradient(gen_loss, generator$variables)
gradients_of_discriminator <-
  disc_tape$gradient(disc_loss, discriminator$variables)
      
generator_optimizer$apply_gradients(purrr::transpose(
  checklist(gradients_of_generator, generator$variables)
))
discriminator_optimizer$apply_gradients(purrr::transpose(
  checklist(gradients_of_discriminator, discriminator$variables)
))
      
total_loss_gen <- total_loss_gen + gen_loss
total_loss_disc <- total_loss_disc + disc_loss

This ends the loop over batches. End off the loop over epochs displaying present losses and saving just a few of the generator’s paintings:

cat("Time for epoch ", epoch, ": ", Sys.time() - begin, "n")
cat("Generator loss: ", total_loss_gen$numpy() / batches_per_epoch, "n")
cat("Discriminator loss: ", total_loss_disc$numpy() / batches_per_epoch, "nn")
if (epoch %% 10 == 0)
  generate_and_save_images(generator,
                           epoch,
                           random_vector_for_generation)

Right here’s the coaching loop once more, proven as an entire – even together with the traces for reporting on progress, it’s remarkably concise, and permits for a fast grasp of what’s going on:

practice <- operate(dataset, epochs, noise_dim) {
  for (epoch in seq_len(num_epochs)) {
    begin <- Sys.time()
    total_loss_gen <- 0
    total_loss_disc <- 0
    iter <- make_iterator_one_shot(train_dataset)
    
    until_out_of_range({
      batch <- iterator_get_next(iter)
      noise <- k_random_normal(c(batch_size, noise_dim))
      with(tf$GradientTape() %as% gen_tape, { with(tf$GradientTape() %as% disc_tape, {
        generated_images <- generator(noise)
        disc_real_output <- discriminator(batch, coaching = TRUE)
        disc_generated_output <-
          discriminator(generated_images, coaching = TRUE)
        gen_loss <- generator_loss(disc_generated_output)
        disc_loss <-
          discriminator_loss(disc_real_output, disc_generated_output)
      }) })
      
      gradients_of_generator <-
        gen_tape$gradient(gen_loss, generator$variables)
      gradients_of_discriminator <-
        disc_tape$gradient(disc_loss, discriminator$variables)
      
      generator_optimizer$apply_gradients(purrr::transpose(
        checklist(gradients_of_generator, generator$variables)
      ))
      discriminator_optimizer$apply_gradients(purrr::transpose(
        checklist(gradients_of_discriminator, discriminator$variables)
      ))
      
      total_loss_gen <- total_loss_gen + gen_loss
      total_loss_disc <- total_loss_disc + disc_loss
      
    })
    
    cat("Time for epoch ", epoch, ": ", Sys.time() - begin, "n")
    cat("Generator loss: ", total_loss_gen$numpy() / batches_per_epoch, "n")
    cat("Discriminator loss: ", total_loss_disc$numpy() / batches_per_epoch, "nn")
    if (epoch %% 10 == 0)
      generate_and_save_images(generator,
                               epoch,
                               random_vector_for_generation)
    
  }
}

Right here’s the operate for saving generated photos…

generate_and_save_images <- operate(mannequin, epoch, test_input) {
  predictions <- mannequin(test_input, coaching = FALSE)
  png(paste0("images_epoch_", epoch, ".png"))
  par(mfcol = c(5, 5))
  par(mar = c(0.5, 0.5, 0.5, 0.5),
      xaxs = 'i',
      yaxs = 'i')
  for (i in 1:25) {
    img <- predictions[i, , , 1]
    img <- t(apply(img, 2, rev))
    picture(
      1:28,
      1:28,
      img * 127.5 + 127.5,
      col = grey((0:255) / 255),
      xaxt = 'n',
      yaxt = 'n'
    )
  }
  dev.off()
}

… and we’re able to go!

num_epochs <- 150
practice(train_dataset, num_epochs, noise_dim)

Outcomes

Listed here are some generated photos after coaching for 150 epochs:

As they are saying, your outcomes will most actually range!

Conclusion

Whereas actually tuning GANs will stay a problem, we hope we had been capable of present that mapping ideas to code isn’t tough when utilizing keen execution. In case you’ve performed round with GANs earlier than, you could have discovered you wanted to pay cautious consideration to arrange the losses the proper manner, freeze the discriminator’s weights when wanted, and many others. This want goes away with keen execution.
In upcoming posts, we’ll present additional examples the place utilizing it makes mannequin growth simpler.