{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"## Import the libraries","metadata":{}},{"cell_type":"code","source":"from kaggle_datasets import KaggleDatasets\nimport matplotlib.pyplot as plt\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport PIL\nimport shutil\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\nimport tensorflow_addons as tfa","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:20:42.730882Z","iopub.execute_input":"2021-12-01T04:20:42.731632Z","iopub.status.idle":"2021-12-01T04:20:42.73695Z","shell.execute_reply.started":"2021-12-01T04:20:42.7316Z","shell.execute_reply":"2021-12-01T04:20:42.736074Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Enable the TPU cluster","metadata":{}},{"cell_type":"code","source":"# Enable the TPU cluster resolver.\ntry:\n    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()\n    print('Device:', tpu.master())\n    tf.config.experimental_connect_to_cluster(tpu)\n    tf.tpu.experimental.initialize_tpu_system(tpu)\n    strategy = tf.distribute.experimental.TPUStrategy(tpu)\nexcept:\n    strategy = tf.distribute.get_strategy()\nprint('Number of replicas:', strategy.num_replicas_in_sync)\n\nAUTOTUNE = tf.data.experimental.AUTOTUNE\n    \nprint(tf.__version__)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:20:49.857401Z","iopub.execute_input":"2021-12-01T04:20:49.858231Z","iopub.status.idle":"2021-12-01T04:20:55.64587Z","shell.execute_reply.started":"2021-12-01T04:20:49.858197Z","shell.execute_reply":"2021-12-01T04:20:55.64528Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Import the Dataset\nRequires turning on Internet on Settings tab to work.","metadata":{}},{"cell_type":"code","source":"GCS_PATH = KaggleDatasets().get_gcs_path()","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:21:08.044664Z","iopub.execute_input":"2021-12-01T04:21:08.04496Z","iopub.status.idle":"2021-12-01T04:21:08.535026Z","shell.execute_reply.started":"2021-12-01T04:21:08.044933Z","shell.execute_reply":"2021-12-01T04:21:08.534046Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"MONET_FILENAMES = tf.io.gfile.glob(str(GCS_PATH + '/monet_tfrec/*.tfrec'))\nprint('Monet TF records: ', len(MONET_FILENAMES))\n\nPHOTO_FILENAMES = tf.io.gfile.glob(str(GCS_PATH + '/photo_tfrec/*.tfrec'))\nprint('Photo TF records: ', len(PHOTO_FILENAMES))","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:21:10.86637Z","iopub.execute_input":"2021-12-01T04:21:10.86664Z","iopub.status.idle":"2021-12-01T04:21:11.035174Z","shell.execute_reply.started":"2021-12-01T04:21:10.866614Z","shell.execute_reply":"2021-12-01T04:21:11.034102Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Image Helper functions","metadata":{}},{"cell_type":"code","source":"IMAGE_SIZE = [256, 256]\n\ndef decode_image(image):\n    image = tf.image.decode_jpeg(image, channels=3)\n    image = (tf.cast(image, tf.float32) / 127.5) - 1\n    image = tf.reshape(image, [*IMAGE_SIZE, 3])\n    return image\n\ndef read_tfrecord(example):\n    tfrecord_format = {\n        \"image_name\": tf.io.FixedLenFeature([], tf.string),\n        \"image\": tf.io.FixedLenFeature([], tf.string),\n        \"target\": tf.io.FixedLenFeature([], tf.string)\n    }\n    example = tf.io.parse_single_example(example, tfrecord_format)\n    image = decode_image(example['image'])\n    return image\n\ndef load_dataset(filenames, labeled=True, ordered=False):\n    dataset = tf.data.TFRecordDataset(filenames)\n    dataset = dataset.map(read_tfrecord, num_parallel_calls=AUTOTUNE)\n    return dataset","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:21:14.659868Z","iopub.execute_input":"2021-12-01T04:21:14.660162Z","iopub.status.idle":"2021-12-01T04:21:14.669808Z","shell.execute_reply.started":"2021-12-01T04:21:14.66013Z","shell.execute_reply":"2021-12-01T04:21:14.668855Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Load the Dataset for the Model","metadata":{}},{"cell_type":"code","source":"monet_ds = load_dataset(MONET_FILENAMES, labeled = True).batch(1)\nphoto_ds = load_dataset(PHOTO_FILENAMES, labeled = True).batch(1)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:29.521103Z","iopub.execute_input":"2021-12-01T04:22:29.521381Z","iopub.status.idle":"2021-12-01T04:22:29.702585Z","shell.execute_reply.started":"2021-12-01T04:22:29.521356Z","shell.execute_reply":"2021-12-01T04:22:29.701819Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Show a sample of the Dataset","metadata":{}},{"cell_type":"code","source":"example_monet = next(iter(monet_ds))\nexample_photo = next(iter(photo_ds))\n\nplt.subplot(1,2,1)\nplt.title('Photo')\nplt.imshow(example_photo[0] *0.5 + 0.5)\n\nplt.subplot(1,2,2)\nplt.title('Monet')\nplt.imshow(example_monet[0]*0.5 + 0.5)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:33.877016Z","iopub.execute_input":"2021-12-01T04:22:33.877456Z","iopub.status.idle":"2021-12-01T04:22:35.208517Z","shell.execute_reply.started":"2021-12-01T04:22:33.877418Z","shell.execute_reply":"2021-12-01T04:22:35.20783Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Helper Functions for the GAN Model","metadata":{}},{"cell_type":"code","source":"# Weights initializer for the layers.\nkernel_init = keras.initializers.RandomNormal(mean = 0.0, stddev = 0.02)\n# Gamma initializer for instance normalization.\ngamma_init = keras.initializers.RandomNormal(mean = 0.0, stddev = 0.02)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:38.526811Z","iopub.execute_input":"2021-12-01T04:22:38.527332Z","iopub.status.idle":"2021-12-01T04:22:38.53142Z","shell.execute_reply.started":"2021-12-01T04:22:38.527299Z","shell.execute_reply":"2021-12-01T04:22:38.530865Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"class ConstantPadding2D(layers.Layer):\n    '''Implements Reflection Padding as a layer.\n\n    Args:\n        padding(tuple): Amount of padding for the\n        spatial dimensions.\n\n    Returns:\n        A padded tensor with the same type as the input tensor.\n    '''\n    \n    def __init__(self, padding = (1,1), **kwargs):\n        self.padding = padding\n        super(ConstantPadding2D, self).__init__(**kwargs)\n        \n    def call(self, input_tensor, mask = None):\n        padding_width, padding_height = self.padding\n        padding_tensor = [\n            [0,0],\n            [padding_height, padding_height],\n            [padding_width, padding_width],\n            [0,0]\n        ]\n        return tf.pad(input_tensor, padding_tensor, mode = 'CONSTANT')\n    \ndef residual_block(\n    x,\n    activation,\n    kernel_initializer = kernel_init,\n    kernel_size = (3,3),\n    strides = (1,1),\n    padding = 'valid',\n    gamma_initializer = gamma_init,\n    use_bias = False,\n):\n    dim = x.shape[-1]\n    input_tensor = x\n    \n    x = ConstantPadding2D()(input_tensor)\n    x = layers.Conv2D(\n        dim,\n        kernel_size,\n        strides = strides,\n        kernel_initializer = kernel_initializer,\n        padding = padding,\n        use_bias = use_bias,\n    )(x)\n    x = tfa.layers.InstanceNormalization(gamma_initializer = gamma_initializer)(x)\n    x = activation(x)\n    x = ConstantPadding2D()(x)\n    x = layers.Conv2D(\n        dim,\n        kernel_size,\n        strides = strides,\n        kernel_initializer = kernel_initializer,\n        padding = padding,\n        use_bias = use_bias,\n    )(x)\n    x = tfa.layers.InstanceNormalization(gamma_initializer = gamma_initializer)(x)\n    x = layers.add([input_tensor, x])\n    return x\n\ndef downsample(\n    x,\n    filters,\n    activation,\n    kernel_initializer = kernel_init,\n    kernel_size = (3,3),\n    strides = (2,2),\n    padding = 'same',\n    gamma_initializer = gamma_init,\n    use_bias = False,\n):\n    x = layers.Conv2D(\n        filters,\n        kernel_size,\n        strides = strides,\n        kernel_initializer = kernel_initializer,\n        padding = padding,\n        use_bias = use_bias,\n    )(x)\n    if activation:\n        x = activation(x)\n    return x\n\ndef upsample(\n    x,\n    filters,\n    activation,\n    kernel_initializer = kernel_init,\n    kernel_size = (3,3),\n    strides = (2,2),\n    padding = 'same',\n    gamma_initializer = gamma_init,\n    use_bias = False,\n):\n    x = layers.Conv2DTranspose(\n        filters,\n        kernel_size,\n        strides = strides,\n        padding = padding,\n        kernel_initializer = kernel_initializer,\n        use_bias = use_bias,\n    )(x)\n    x = tfa.layers.InstanceNormalization(gamma_initializer = gamma_initializer)(x)\n    if activation:\n        x = activation(x)\n    return x","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:41.785788Z","iopub.execute_input":"2021-12-01T04:22:41.786301Z","iopub.status.idle":"2021-12-01T04:22:41.805008Z","shell.execute_reply.started":"2021-12-01T04:22:41.786255Z","shell.execute_reply":"2021-12-01T04:22:41.804114Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Build the Generator and Discriminator Models","metadata":{}},{"cell_type":"code","source":"#-------------- Generator Model\ndef resnet_generator(filters = 64, num_downsampling_blocks = 2, num_residual_blocks = 9,\n                     num_upsampling_blocks = 2, gamma_initializer = gamma_init, name = None):\n    img_input = layers.Input(shape = (256,256,3), name = name + '_img_input')\n    x = ConstantPadding2D(padding = (3,3))(img_input)\n    x = layers.Conv2D(filters, (7,7), kernel_initializer = kernel_init, use_bias = False)(x)\n    x = tfa.layers.InstanceNormalization(gamma_initializer = gamma_initializer)(x)\n    x = layers.Activation('relu')(x)\n    \n    #Downsampling\n    for _ in range(num_downsampling_blocks):\n        filters *= 2\n        x = downsample(x, filters = filters, activation = layers.Activation('relu'))\n        \n    #Residual blocks\n    for _ in range(num_residual_blocks):\n        x = residual_block(x, activation = layers.Activation('relu'))\n        \n    #Upsampling\n    for _ in range(num_upsampling_blocks):\n        filters //= 2\n        x = upsample(x, filters = filters, activation = layers.Activation('relu'))\n        \n    #final block\n    x = ConstantPadding2D(padding = (3,3))(x)\n    x = layers.Conv2D(3, (7,7), padding = 'valid')(x)\n    x = layers.Activation('tanh')(x)\n    \n    model = keras.models.Model(img_input, x, name = name)\n    return model\n\n\n#-------------- Discriminator Model\ndef discriminator(filters = 64, kernel_initializer = kernel_init, num_downsampling = 3, name = None):\n    img_input = layers.Input(shape = (256,256,3), name = name + '_img_input')\n    x = layers.Conv2D(\n        filters,\n        (4,4),\n        strides = (2,2),\n        padding = 'same',\n        kernel_initializer = kernel_initializer,\n        )(img_input)\n    x = layers.LeakyReLU(0.2)(x)\n    \n    num_filters = filters\n    for num_downsample_block in range(3):\n        num_filters *= 2\n        if num_downsample_block < 2:\n            x = downsample(\n                x,\n                filters = num_filters,\n                activation = layers.LeakyReLU(0.2),\n                kernel_size = (4,4),\n                strides = (2,2),\n            )\n        else:\n            x = downsample(\n                x,\n                filters = num_filters,\n                activation = layers.LeakyReLU(0.2),\n                kernel_size = (4,4),\n                strides = (1,1),\n            )\n    x = layers.Conv2D(1, (4,4), strides = (1,1), padding = 'same', kernel_initializer = kernel_initializer)(x)\n    model = keras.models.Model(img_input, x, name = name)\n    return model","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:46.98641Z","iopub.execute_input":"2021-12-01T04:22:46.986705Z","iopub.status.idle":"2021-12-01T04:22:47.003497Z","shell.execute_reply.started":"2021-12-01T04:22:46.986676Z","shell.execute_reply":"2021-12-01T04:22:47.002883Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Enable TPU on the Generator and Discriminator Models","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    monet_generator = resnet_generator(name = 'generator_monet')  #transforms photos to monet\n    photo_generator = resnet_generator(name = 'generator_photo')  #transforms monet to photos\n    monet_discriminator = discriminator(name = 'discriminator_monet') #differentiates between real monets and generated monets\n    photo_discriminator = discriminator(name = 'discriminator_photo') #differentiates between real photos and generated photos","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:22:51.492532Z","iopub.execute_input":"2021-12-01T04:22:51.493063Z","iopub.status.idle":"2021-12-01T04:22:58.492418Z","shell.execute_reply.started":"2021-12-01T04:22:51.493032Z","shell.execute_reply":"2021-12-01T04:22:58.491542Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"to_monet = monet_generator(example_photo)\n\nplt.subplot(1,2,1)\nplt.title('Original Photo')\nplt.imshow(example_photo[0]*0.5 + 0.5)\n\nplt.subplot(1,2,2)\nplt.title('Generated Monet')\nplt.imshow(to_monet[0]*0.5 + 0.5)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:00.830765Z","iopub.execute_input":"2021-12-01T04:23:00.831202Z","iopub.status.idle":"2021-12-01T04:23:01.55053Z","shell.execute_reply.started":"2021-12-01T04:23:00.831176Z","shell.execute_reply":"2021-12-01T04:23:01.549719Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"We can see that the model is initialized and ready to be trained.","metadata":{}},{"cell_type":"markdown","source":"### Build the Loss Functions to be ran on the TPU","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    def discriminator_loss(real, generated):\n        real_loss = keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.ones_like(real), real)\n        fake_loss = keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.zeros_like(generated), generated)\n        total_disc_loss = real_loss + fake_loss\n        return total_disc_loss*0.5     \n\nwith strategy.scope():\n    def generator_loss(generated):\n        return keras.losses.BinaryCrossentropy(from_logits=True, reduction=tf.keras.losses.Reduction.NONE)(tf.ones_like(generated), generated)\n\nwith strategy.scope():\n    def cycle_loss(real_image, cycled_image, LAMBDA):\n        loss = tf.reduce_mean(tf.abs(real_image - cycled_image))\n        return loss*LAMBDA\n\nwith strategy.scope():\n    def identity_loss(real_image, same_image, LAMBDA1, LAMBDA2):\n        loss = tf.reduce_mean(tf.abs(real_image - same_image))\n        return loss*LAMBDA1*LAMBDA2","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:05.642979Z","iopub.execute_input":"2021-12-01T04:23:05.643252Z","iopub.status.idle":"2021-12-01T04:23:05.653213Z","shell.execute_reply.started":"2021-12-01T04:23:05.643224Z","shell.execute_reply":"2021-12-01T04:23:05.652121Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Configure the Adam Optimizers to be ran on the TPU","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    m_gen_optimizer = keras.optimizers.Adam(learning_rate = 2e-4, beta_1 = 0.5)\n    p_gen_optimizer = keras.optimizers.Adam(learning_rate = 2e-4, beta_1 = 0.5)\n    m_disc_optimizer = keras.optimizers.Adam(learning_rate = 2e-4, beta_1 = 0.5)\n    p_disc_optimizer = keras.optimizers.Adam(learning_rate = 2e-4, beta_1 = 0.5)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:08.666179Z","iopub.execute_input":"2021-12-01T04:23:08.666473Z","iopub.status.idle":"2021-12-01T04:23:08.672928Z","shell.execute_reply.started":"2021-12-01T04:23:08.666443Z","shell.execute_reply":"2021-12-01T04:23:08.672068Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Build the CycleGAN Model","metadata":{}},{"cell_type":"code","source":"class CycleGAN(keras.Model):\n    def __init__(self, monet_generator, photo_generator,\n                 monet_discriminator, photo_discriminator,\n                 lambda_cycle = 10.0, lambda_identity = 0.5):\n        super(CycleGAN, self).__init__()\n        self.m_gen = monet_generator\n        self.p_gen = photo_generator\n        self.m_disc = monet_discriminator\n        self.p_disc = photo_discriminator\n        self.lambda_cycle = lambda_cycle\n        self.lambda_identity = lambda_identity\n        \n    def compile(self, m_gen_optimizer, p_gen_optimizer,\n                m_disc_optimizer,p_disc_optimizer,\n                gen_loss_fn, disc_loss_fn, cycle_loss_fn, identity_loss_fn):\n        super(CycleGAN, self).compile()\n        self.m_gen_optimizer = m_gen_optimizer\n        self.p_gen_optimizer = p_gen_optimizer\n        self.m_disc_optimizer = m_disc_optimizer\n        self.p_disc_optimizer = p_disc_optimizer\n        self.gen_loss_fn = gen_loss_fn\n        self.disc_loss_fn = disc_loss_fn\n        self.cycle_loss_fn = cycle_loss_fn\n        self.identity_loss_fn = identity_loss_fn\n    \n    def train_step(self, batch):\n        real_monet, real_photo = batch\n        with tf.GradientTape(persistent = True) as tape:\n            #photo --> monet --> photo\n            fake_monet = self.m_gen(real_photo, training = True)\n            cycled_photo = self.p_gen(fake_monet, training = True)\n            \n            #monet --> photo --> monet\n            fake_photo = self.p_gen(real_monet, training = True)\n            cycled_monet = self.m_gen(fake_photo, training = True)\n            \n            #identity mapping\n            same_monet = self.m_gen(real_monet, training = True)\n            same_photo = self.p_gen(real_photo, training = True)\n            \n            #discriminator output\n            disc_real_monet = self.m_disc(real_monet, training = True)\n            disc_fake_monet = self.m_disc(fake_monet, training = True)\n            \n            disc_real_photo = self.p_disc(real_photo, training = True)\n            disc_fake_photo = self.p_disc(fake_photo, training = True)\n            \n            #generator cycle losses\n            cycled_loss_monet = self.cycle_loss_fn(real_monet, cycled_monet, self.lambda_cycle)\n            cycled_loss_photo = self.cycle_loss_fn(real_photo, cycled_photo, self.lambda_cycle)\n            total_cycle_loss = cycled_loss_monet + cycled_loss_photo\n            \n            #generator losses\n            monet_gen_loss = self.gen_loss_fn(disc_fake_monet)\n            photo_gen_loss = self.gen_loss_fn(disc_fake_photo)\n            \n            #identity losses\n            monet_id_loss = self.identity_loss_fn(real_monet, same_monet, self.lambda_cycle, self.lambda_identity)\n            photo_id_loss = self.identity_loss_fn(real_photo, same_photo, self.lambda_cycle, self.lambda_identity)\n            total_id_loss = monet_id_loss + photo_id_loss\n            \n            #total generator loss\n            total_monet_gen_loss = monet_gen_loss + total_cycle_loss + total_id_loss\n            total_photo_gen_loss = photo_gen_loss + total_cycle_loss + total_id_loss\n            \n            #discriminator losses\n            monet_disc_loss = self.disc_loss_fn(disc_real_monet, disc_fake_monet)\n            photo_disc_loss = self.disc_loss_fn(disc_real_photo, disc_fake_photo)\n        \n        #calculate gradients\n        monet_generator_gradients = tape.gradient(total_monet_gen_loss, self.m_gen.trainable_variables)\n        photo_generator_gradients = tape.gradient(total_photo_gen_loss, self.p_gen.trainable_variables)\n        \n        monet_discriminator_gradients = tape.gradient(monet_disc_loss, self.m_disc.trainable_variables)\n        photo_discriminator_gradients = tape.gradient(photo_disc_loss, self.p_disc.trainable_variables)\n        \n        self.m_gen_optimizer.apply_gradients(zip(monet_generator_gradients, self.m_gen.trainable_variables))\n        self.p_gen_optimizer.apply_gradients(zip(photo_generator_gradients, self.p_gen.trainable_variables))\n        self.m_disc_optimizer.apply_gradients(zip(monet_discriminator_gradients, self.m_disc.trainable_variables))\n        self.p_disc_optimizer.apply_gradients(zip(photo_discriminator_gradients, self.p_disc.trainable_variables))\n        \n        return {\n            'monet_gen_loss': total_monet_gen_loss,\n            'photo_gen_loss': total_photo_gen_loss,\n            'monet_disc_loss': monet_disc_loss,\n            'photo_disc_loss': photo_disc_loss\n        }","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:12.088691Z","iopub.execute_input":"2021-12-01T04:23:12.08916Z","iopub.status.idle":"2021-12-01T04:23:12.109218Z","shell.execute_reply.started":"2021-12-01T04:23:12.089118Z","shell.execute_reply":"2021-12-01T04:23:12.108097Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Compile the CycleGAN Model to run on the TPU","metadata":{}},{"cell_type":"code","source":"with strategy.scope():\n    cycle_gan_model = CycleGAN(monet_generator, photo_generator, monet_discriminator, photo_discriminator)\n    cycle_gan_model.compile(m_gen_optimizer = m_gen_optimizer, p_gen_optimizer = p_gen_optimizer,\n                            m_disc_optimizer = m_disc_optimizer, p_disc_optimizer = p_disc_optimizer,\n                            gen_loss_fn = generator_loss, disc_loss_fn = discriminator_loss,\n                            cycle_loss_fn = cycle_loss, identity_loss_fn = identity_loss)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:44.671817Z","iopub.execute_input":"2021-12-01T04:23:44.672555Z","iopub.status.idle":"2021-12-01T04:23:44.725725Z","shell.execute_reply.started":"2021-12-01T04:23:44.672503Z","shell.execute_reply":"2021-12-01T04:23:44.724919Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Train the CycleGAN Model with 50 epochs\nThe model fitting will be stuck at Epoch 1/50 for sometime while loading to the TPU, after sometime, the TPU will kickstart and the training will begin.\n\nWith the TPU, each epoch takes approximately 2 minutes. This means 50 epochs takes about 120 minutes or about 2 hours.","metadata":{}},{"cell_type":"code","source":"history = cycle_gan_model.fit(tf.data.Dataset.zip((monet_ds, photo_ds)), epochs = 50)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T04:23:52.845926Z","iopub.execute_input":"2021-12-01T04:23:52.846775Z","iopub.status.idle":"2021-12-01T05:47:51.530788Z","shell.execute_reply.started":"2021-12-01T04:23:52.846737Z","shell.execute_reply":"2021-12-01T05:47:51.530013Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"### Finally, Visualize the Predictions from the monet_generator","metadata":{}},{"cell_type":"code","source":"_, ax = plt.subplots(6,2,figsize = (20,20))\nfor i, img in enumerate(photo_ds.take(6)):\n    prediction = monet_generator(img, training = False)[0].numpy()\n    prediction = prediction*0.5 + 0.5\n    img = (img[0]*0.5 + 0.5).numpy()\n    ax[i, 0].imshow(img)\n    ax[i, 1].imshow(prediction)\n    ax[i, 0].set_title('Input Photo')\n    ax[i, 1].set_title('Generated Photo')\n    ax[i, 0].axis('off')\n    ax[i, 1].axis('off')\nplt.tight_layout()\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2021-12-01T05:47:57.893305Z","iopub.execute_input":"2021-12-01T05:47:57.893579Z","iopub.status.idle":"2021-12-01T05:48:01.079419Z","shell.execute_reply.started":"2021-12-01T05:47:57.893545Z","shell.execute_reply":"2021-12-01T05:48:01.078517Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"","metadata":{}},{"cell_type":"markdown","source":"### Generate the Kaggle Submission\nFirst make a directory to store the images from the model.","metadata":{}},{"cell_type":"code","source":"# Make an images directory.\n! mkdir ../images","metadata":{"execution":{"iopub.status.busy":"2021-12-01T05:48:01.080582Z","iopub.execute_input":"2021-12-01T05:48:01.08082Z","iopub.status.idle":"2021-12-01T05:48:01.929982Z","shell.execute_reply.started":"2021-12-01T05:48:01.080795Z","shell.execute_reply":"2021-12-01T05:48:01.928817Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"##### Obtain the monet generated outputs from monet_generator and save them into the images folder. Finally, zip it.","metadata":{}},{"cell_type":"code","source":"records_n = sum(1 for record in photo_ds)\nprint(\"# of photos to predict: %s\" % records_n)","metadata":{"execution":{"iopub.status.busy":"2021-12-01T06:20:13.225176Z","iopub.execute_input":"2021-12-01T06:20:13.225444Z","iopub.status.idle":"2021-12-01T06:20:22.300308Z","shell.execute_reply.started":"2021-12-01T06:20:13.225418Z","shell.execute_reply":"2021-12-01T06:20:22.299472Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"with strategy.scope():\n    i = 1\n    for img in photo_ds:\n        prediction = monet_generator(img, training=False)[0].numpy()\n        prediction = (prediction*127.5 + 127.5).astype(np.uint8)\n        im = PIL.Image.fromarray(prediction)\n        im.save(\"../images/\" + str(i) + \".jpg\")\n        i += 1\n        if (i%101 == 0):\n            print(\"Generated Prediction for photo: %s/%s\" % (i, records_n))\n    \n# Make an archive in /kaggle/working/images as images.zip\nshutil.make_archive(\"/kaggle/working/images\", 'zip', \"/kaggle/images\")","metadata":{"execution":{"iopub.status.busy":"2021-12-01T06:26:30.759714Z","iopub.execute_input":"2021-12-01T06:26:30.760006Z","iopub.status.idle":"2021-12-01T07:07:48.017025Z","shell.execute_reply.started":"2021-12-01T06:26:30.759979Z","shell.execute_reply":"2021-12-01T07:07:48.016004Z"},"trusted":true},"execution_count":null,"outputs":[]}]}