{"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":"# **Imports**","metadata":{"id":"metVEueVcq6V"}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nimport matplotlib.pyplot as plt\n\nimport tensorflow as tf\nfrom tensorflow import keras\n\nfrom keras import Model, layers, losses, optimizers, initializers\nfrom keras.layers import Conv2D, Conv2DTranspose\nfrom keras.models import Sequential","metadata":{"id":"Px4ES4XX3RwB","execution":{"iopub.status.busy":"2021-07-21T18:30:59.719793Z","iopub.execute_input":"2021-07-21T18:30:59.720503Z","iopub.status.idle":"2021-07-21T18:31:05.922885Z","shell.execute_reply.started":"2021-07-21T18:30:59.720380Z","shell.execute_reply":"2021-07-21T18:31:05.921816Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Load Data**","metadata":{"id":"ufjRiUGEdEDq"}},{"cell_type":"code","source":"MONET_FILENAMES = tf.io.gfile.glob('/kaggle/input/gan-getting-started/monet_tfrec/monet*.tfrec')\nPHOTO_FILENAMES = tf.io.gfile.glob('/kaggle/input/gan-getting-started/photo_tfrec/photo*.tfrec')\n\nprint('Monet Number Of Files:', len(MONET_FILENAMES))\nprint('Photo Number Of Files:', len(PHOTO_FILENAMES))","metadata":{"id":"NvK3EQ-B3oBb","outputId":"3d383892-ae4e-4b8d-f5ff-29488cf83351","execution":{"iopub.status.busy":"2021-07-21T18:31:05.924272Z","iopub.execute_input":"2021-07-21T18:31:05.924559Z","iopub.status.idle":"2021-07-21T18:31:05.955216Z","shell.execute_reply.started":"2021-07-21T18:31:05.924531Z","shell.execute_reply":"2021-07-21T18:31:05.954142Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"AUTOTUNE = tf.data.experimental.AUTOTUNE\nIMAGE_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):\n    dataset = tf.data.TFRecordDataset(filenames)\n    dataset = dataset.map(read_tfrecord, num_parallel_calls=AUTOTUNE)\n    return dataset","metadata":{"id":"EXrR2bim7TzL","execution":{"iopub.status.busy":"2021-07-21T18:31:05.957139Z","iopub.execute_input":"2021-07-21T18:31:05.957421Z","iopub.status.idle":"2021-07-21T18:31:05.964910Z","shell.execute_reply.started":"2021-07-21T18:31:05.957393Z","shell.execute_reply":"2021-07-21T18:31:05.963803Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_ds = load_dataset(MONET_FILENAMES).batch(1)\nphoto_ds = load_dataset(PHOTO_FILENAMES).batch(1)","metadata":{"id":"1C9GB5_k5s48","execution":{"iopub.status.busy":"2021-07-21T18:31:05.966372Z","iopub.execute_input":"2021-07-21T18:31:05.966651Z","iopub.status.idle":"2021-07-21T18:31:06.182198Z","shell.execute_reply.started":"2021-07-21T18:31:05.966625Z","shell.execute_reply":"2021-07-21T18:31:06.181307Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Visualize sample**","metadata":{"id":"d2AO7axRi4Tv"}},{"cell_type":"code","source":"example_monet = next(iter(monet_ds))\nexample_photo = next(iter(photo_ds))\n\nplt.subplot(121)\nplt.title('Photo')\nplt.imshow(example_photo[0] * 0.5 + 0.5)\n\nplt.subplot(122)\nplt.title('Monet')\nplt.imshow(example_monet[0] * 0.5 + 0.5)","metadata":{"id":"16V3v5ay8bDL","outputId":"5e84a37e-a21f-4348-d954-73fd20c68767","execution":{"iopub.status.busy":"2021-07-21T18:31:06.183370Z","iopub.execute_input":"2021-07-21T18:31:06.183635Z","iopub.status.idle":"2021-07-21T18:31:06.684703Z","shell.execute_reply.started":"2021-07-21T18:31:06.183608Z","shell.execute_reply":"2021-07-21T18:31:06.683564Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Downsample & Upsample**","metadata":{"id":"iJ-S9oVEi6DL"}},{"cell_type":"code","source":"OUTPUT_CHANNELS = 3","metadata":{"id":"1hiCr0V7kj5x","execution":{"iopub.status.busy":"2021-07-21T18:31:06.686269Z","iopub.execute_input":"2021-07-21T18:31:06.686863Z","iopub.status.idle":"2021-07-21T18:31:06.691488Z","shell.execute_reply.started":"2021-07-21T18:31:06.686821Z","shell.execute_reply":"2021-07-21T18:31:06.690206Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def downsample(filters, size):\n    initializer = tf.random_normal_initializer(0., 0.03)\n    \n    model = Sequential()\n    model.add(Conv2D(filters, size, strides=2, padding='same', kernel_initializer=initializer, use_bias=False))\n    model.add(tf.keras.layers.LeakyReLU())\n\n    return model","metadata":{"id":"HZw-f-Jr86Mw","execution":{"iopub.status.busy":"2021-07-21T18:31:06.693060Z","iopub.execute_input":"2021-07-21T18:31:06.693512Z","iopub.status.idle":"2021-07-21T18:31:06.705353Z","shell.execute_reply.started":"2021-07-21T18:31:06.693482Z","shell.execute_reply":"2021-07-21T18:31:06.704601Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def upsample(filters, size):\n    initializer = tf.random_normal_initializer(0., 0.03)\n\n    model = Sequential()\n    model.add(Conv2DTranspose(filters, size, activation='relu', strides=2, padding='same', kernel_initializer=initializer, use_bias=False))\n    \n    return model","metadata":{"id":"1aZ8PKFzqHA4","execution":{"iopub.status.busy":"2021-07-21T18:31:06.707306Z","iopub.execute_input":"2021-07-21T18:31:06.707764Z","iopub.status.idle":"2021-07-21T18:31:06.718258Z","shell.execute_reply.started":"2021-07-21T18:31:06.707735Z","shell.execute_reply":"2021-07-21T18:31:06.717560Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Create generator & discriminator**","metadata":{"id":"O9u6VU6LiILw"}},{"cell_type":"code","source":"def Generator():\n    inputs = layers.Input(shape=[256,256,3])\n\n    down_stack = [\n        downsample(64, 4), # (128, 128, 64)\n        downsample(128, 4), # (64, 64, 128)\n        downsample(256, 4), # (32, 32, 256)\n        downsample(512, 4), # (16, 16, 512)\n        downsample(512, 4), # (8, 8, 512)\n        downsample(512, 4), # (4, 4, 512)\n        downsample(512, 4), # (2, 2, 512)\n        downsample(1024, 4), # (1, 1, 512)\n    ]\n\n    up_stack = [\n        upsample(1024, 4), # (2, 2, 1024)\n        upsample(512, 4), # (4, 4, 1024)\n        upsample(512, 4), # (8, 8, 1024)\n        upsample(512, 4), # (16, 16, 1024)\n        upsample(256, 4), # (32, 32, 512)\n        upsample(128, 4), # (64, 64, 256)\n        upsample(64, 4), # (128, 128, 128)\n    ]\n\n    initializer = tf.random_normal_initializer(0., 0.03)\n    last = Conv2DTranspose(OUTPUT_CHANNELS, 4, strides=2, padding='same', kernel_initializer=initializer, activation='tanh') # (256, 256, 3)\n    \n    x = inputs\n\n    # Downsampling through the model\n    skips = []\n    for down in down_stack:\n        x = down(x)\n        skips.append(x)\n\n    skips = reversed(skips[:-1])\n\n    # Upsampling and establishing the skip connections\n    for up, skip in zip(up_stack, skips):\n        x = up(x)\n        x = layers.Concatenate()([x, skip])\n\n    x = last(x)\n\n    return Model(inputs=inputs, outputs=x)","metadata":{"id":"q37H0KIG9Kq4","execution":{"iopub.status.busy":"2021-07-21T18:31:06.719825Z","iopub.execute_input":"2021-07-21T18:31:06.720099Z","iopub.status.idle":"2021-07-21T18:31:06.736573Z","shell.execute_reply.started":"2021-07-21T18:31:06.720073Z","shell.execute_reply":"2021-07-21T18:31:06.735591Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def Discriminator():\n    initializer = tf.random_normal_initializer(0., 0.03)\n\n    discInputs = layers.Input(shape=[256, 256, 3], name='input_image')\n    x = discInputs\n\n    firstLayer = downsample(64, 4)(x) # (128, 128, 64)\n    secondLayer = downsample(128, 4)(firstLayer) # (64, 64, 128)\n    thirdLayer = downsample(256, 4)(secondLayer) # (32, 32, 256)\n\n    firstZeroPadLayer = layers.ZeroPadding2D()(thirdLayer) # (34, 34, 256)\n    convLayer = Conv2D(512, 4, strides=1, kernel_initializer=initializer, use_bias=False)(firstZeroPadLayer) # (31, 31, 512)\n\n    leaky_relu = layers.LeakyReLU()(convLayer)\n\n    secondZeroPadLayer = layers.ZeroPadding2D()(leaky_relu) # (33, 33, 512)\n\n    lastLayer = layers.Conv2D(1, 4, strides=1, kernel_initializer=initializer)(secondZeroPadLayer) # (30, 30, 1)\n\n    return Model(inputs=discInputs, outputs=lastLayer)","metadata":{"id":"6hhoamgG9Uw7","execution":{"iopub.status.busy":"2021-07-21T18:31:06.737962Z","iopub.execute_input":"2021-07-21T18:31:06.738389Z","iopub.status.idle":"2021-07-21T18:31:06.748889Z","shell.execute_reply.started":"2021-07-21T18:31:06.738283Z","shell.execute_reply":"2021-07-21T18:31:06.747926Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"monet_generator = Generator() # transforms photos to Monet-esque paintings\nphoto_generator = Generator() # transforms Monet paintings to be more like photos\n\nmonet_discriminator = Discriminator() # differentiates real Monet paintings and generated Monet paintings\nphoto_discriminator = Discriminator() # differentiates real photos and generated photos","metadata":{"id":"8dkTP3Cc9avH","execution":{"iopub.status.busy":"2021-07-21T18:31:06.750365Z","iopub.execute_input":"2021-07-21T18:31:06.750943Z","iopub.status.idle":"2021-07-21T18:31:09.497578Z","shell.execute_reply.started":"2021-07-21T18:31:06.750898Z","shell.execute_reply":"2021-07-21T18:31:09.496819Z"},"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(\"Monet-esque Photo\")\nplt.imshow(to_monet[0] * 0.5 + 0.5)\nplt.show()","metadata":{"id":"TpTlIwlnHyTa","outputId":"85141efe-5d2f-4022-a5fa-56cf995a0316","execution":{"iopub.status.busy":"2021-07-21T18:31:09.498637Z","iopub.execute_input":"2021-07-21T18:31:09.499067Z","iopub.status.idle":"2021-07-21T18:31:10.278758Z","shell.execute_reply.started":"2021-07-21T18:31:09.499026Z","shell.execute_reply":"2021-07-21T18:31:10.278035Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Define the CycleGan**","metadata":{"id":"HAhjRBDPiePh"}},{"cell_type":"code","source":"class CycleGan(Model):\n    def __init__(self, monet_generator, photo_generator, monet_discriminator, photo_discriminator, lambda_cycle=10):\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        \n    def compile(self, m_gen_opt, p_gen_opt, m_disc_opt, p_disc_opt, gen_loss_fn, disc_loss_fn, cycle_loss_fn, identity_loss_fn):\n        super(CycleGan, self).compile()\n        self.m_gen_opt = m_gen_opt\n        self.p_gen_opt = p_gen_opt\n        self.m_disc_opt = m_disc_opt\n        self.p_disc_opt = p_disc_opt\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_data):\n        real_monet, real_photo = batch_data\n        \n        with tf.GradientTape(persistent=True) as tape:\n            # Generate photo to monet and then generate back to photo\n            fake_monet = self.m_gen(real_photo, training=True)\n            cycled_photo = self.p_gen(fake_monet, training=True)\n\n            # Generate monet to photo and then generate back to monet\n            fake_photo = self.p_gen(real_monet, training=True)\n            cycled_monet = self.m_gen(fake_photo, training=True)\n\n            # generating itself\n            same_monet = self.m_gen(real_monet, training=True)\n            same_photo = self.p_gen(real_photo, training=True)\n\n            # Send the real images to the discriminator\n            disc_real_monet = self.m_disc(real_monet, training=True)\n            disc_real_photo = self.p_disc(real_photo, training=True)\n\n            # Send the fake images to the discriminator\n            disc_fake_monet = self.m_disc(fake_monet, training=True)\n            disc_fake_photo = self.p_disc(fake_photo, training=True)\n\n            # evaluates generator loss\n            monet_gen_loss = self.gen_loss_fn(disc_fake_monet)\n            photo_gen_loss = self.gen_loss_fn(disc_fake_photo)\n\n            # evaluates total cycle consistency loss\n            total_cycle_loss = self.cycle_loss_fn(real_monet, cycled_monet) + self.cycle_loss_fn(real_photo, cycled_photo)\n\n            # evaluates total generator loss\n            total_monet_gen_loss = monet_gen_loss + self.lambda_cycle * total_cycle_loss + self.identity_loss_fn(real_monet, same_monet)\n            total_photo_gen_loss = photo_gen_loss + self.lambda_cycle * total_cycle_loss + self.identity_loss_fn(real_photo, same_photo)\n\n            # evaluates discriminator loss\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 the gradients for generator and discriminator\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        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        # Apply the gradients to the optimizer\n        self.m_gen_opt.apply_gradients(zip(monet_generator_gradients, self.m_gen.trainable_variables))\n        self.p_gen_opt.apply_gradients(zip(photo_generator_gradients, self.p_gen.trainable_variables))\n        self.m_disc_opt.apply_gradients(zip(monet_discriminator_gradients, self.m_disc.trainable_variables))\n        self.p_disc_opt.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":{"id":"jfpjHR5fH-sY","execution":{"iopub.status.busy":"2021-07-21T18:31:10.279990Z","iopub.execute_input":"2021-07-21T18:31:10.280436Z","iopub.status.idle":"2021-07-21T18:31:10.293618Z","shell.execute_reply.started":"2021-07-21T18:31:10.280391Z","shell.execute_reply":"2021-07-21T18:31:10.292788Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Loss Functions**","metadata":{"id":"jLPczmWik3Ni"}},{"cell_type":"code","source":"def discriminator_loss(real, generated):\n    real_loss = losses.BinaryCrossentropy(from_logits=True, reduction=keras.losses.Reduction.NONE)(tf.ones_like(real), real)\n    generated_loss = losses.BinaryCrossentropy(from_logits=True, reduction=keras.losses.Reduction.NONE)(tf.zeros_like(generated), generated)\n\n    total_disc_loss = real_loss + generated_loss\n\n    return total_disc_loss * 0.5","metadata":{"id":"UbsX9ghCIIDg","execution":{"iopub.status.busy":"2021-07-21T18:31:10.294920Z","iopub.execute_input":"2021-07-21T18:31:10.295240Z","iopub.status.idle":"2021-07-21T18:31:10.310817Z","shell.execute_reply.started":"2021-07-21T18:31:10.295197Z","shell.execute_reply":"2021-07-21T18:31:10.309910Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def generator_loss(generated):\n    return losses.BinaryCrossentropy(from_logits=True, reduction=keras.losses.Reduction.NONE)(tf.ones_like(generated), generated)","metadata":{"id":"NGVM2MdZoOw6","execution":{"iopub.status.busy":"2021-07-21T18:31:10.312382Z","iopub.execute_input":"2021-07-21T18:31:10.313011Z","iopub.status.idle":"2021-07-21T18:31:10.327256Z","shell.execute_reply.started":"2021-07-21T18:31:10.312966Z","shell.execute_reply":"2021-07-21T18:31:10.326120Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def calc_cycle_loss(real_image, cycled_image):\n    return tf.reduce_mean(tf.abs(cycled_image - real_image))","metadata":{"id":"1AhWpXyPIK9o","execution":{"iopub.status.busy":"2021-07-21T18:31:10.328627Z","iopub.execute_input":"2021-07-21T18:31:10.328948Z","iopub.status.idle":"2021-07-21T18:31:10.339613Z","shell.execute_reply.started":"2021-07-21T18:31:10.328918Z","shell.execute_reply":"2021-07-21T18:31:10.338677Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"def identity_loss(real_image, same_image):\n    loss = tf.reduce_mean(tf.abs(real_image - same_image))\n    \n    return 0.5174 * loss","metadata":{"id":"H0zNrPuNIRUR","execution":{"iopub.status.busy":"2021-07-21T18:31:10.340693Z","iopub.execute_input":"2021-07-21T18:31:10.341109Z","iopub.status.idle":"2021-07-21T18:31:10.352834Z","shell.execute_reply.started":"2021-07-21T18:31:10.341078Z","shell.execute_reply":"2021-07-21T18:31:10.351657Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Optimization**","metadata":{"id":"6g1ot28ulPCw"}},{"cell_type":"code","source":"monet_generator_optimizer = optimizers.Adam(2e-4, beta_1=0.5)\nphoto_generator_optimizer = optimizers.Adam(2e-4, beta_1=0.5)\n\nmonet_discriminator_optimizer = optimizers.Adam(2e-4, beta_1=0.5)\nphoto_discriminator_optimizer = optimizers.Adam(2e-4, beta_1=0.5)","metadata":{"id":"8HvON7KXIVja","execution":{"iopub.status.busy":"2021-07-21T18:31:10.354138Z","iopub.execute_input":"2021-07-21T18:31:10.354426Z","iopub.status.idle":"2021-07-21T18:31:10.365798Z","shell.execute_reply.started":"2021-07-21T18:31:10.354398Z","shell.execute_reply":"2021-07-21T18:31:10.364642Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Compilation**","metadata":{"id":"KJuStCMmmDtl"}},{"cell_type":"code","source":"cycle_gan_model = CycleGan(monet_generator, photo_generator, monet_discriminator, photo_discriminator)\n\ncycle_gan_model.compile(\n    m_gen_opt = monet_generator_optimizer,\n    p_gen_opt = photo_generator_optimizer,\n    m_disc_opt = monet_discriminator_optimizer,\n    p_disc_opt = photo_discriminator_optimizer,\n    gen_loss_fn = generator_loss,\n    disc_loss_fn = discriminator_loss,\n    cycle_loss_fn = calc_cycle_loss,\n    identity_loss_fn = identity_loss\n)","metadata":{"id":"llAFjCEnIazw","execution":{"iopub.status.busy":"2021-07-21T18:31:10.367210Z","iopub.execute_input":"2021-07-21T18:31:10.367811Z","iopub.status.idle":"2021-07-21T18:31:10.397579Z","shell.execute_reply.started":"2021-07-21T18:31:10.367757Z","shell.execute_reply":"2021-07-21T18:31:10.396846Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Output**","metadata":{"id":"slOXowZimJzi"}},{"cell_type":"code","source":"def printResults():\n  _, ax = plt.subplots(5, 2, figsize=(15, 15))\n  for i, img in enumerate(photo_ds.shuffle(300).take(5)):\n      prediction = monet_generator(img, training=False)[0].numpy()\n      prediction = (prediction * 127.5 + 127.5).astype(np.uint8)\n      img = (img[0] * 127.5 + 127.5).numpy().astype(np.uint8)\n\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(\"Monet-esque\")\n      ax[i, 0].axis(\"off\")\n      ax[i, 1].axis(\"off\")\n  plt.show()","metadata":{"id":"AtLjBQTbIf5k","execution":{"iopub.status.busy":"2021-07-21T18:31:10.398575Z","iopub.execute_input":"2021-07-21T18:31:10.399017Z","iopub.status.idle":"2021-07-21T18:31:10.405926Z","shell.execute_reply.started":"2021-07-21T18:31:10.398973Z","shell.execute_reply":"2021-07-21T18:31:10.404596Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# **Training**","metadata":{"id":"kQzm0r_5SVxK"}},{"cell_type":"code","source":"history = cycle_gan_model.fit(tf.data.Dataset.zip((monet_ds, photo_ds)), epochs=30)","metadata":{},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"printResults()","metadata":{"execution":{"iopub.status.busy":"2021-07-21T18:31:10.429860Z","iopub.status.idle":"2021-07-21T18:31:10.430259Z"},"trusted":true},"execution_count":null,"outputs":[]}]}