Face Generation

In this project, you'll use generative adversarial networks to generate new images of faces.

Get the Data

You'll be using two datasets in this project:

  • MNIST
  • CelebA

Since the celebA dataset is complex and you're doing GANs in a project for the first time, we want you to test your neural network on MNIST before CelebA. Running the GANs on MNIST will allow you to see how well your model trains sooner.

If you're using FloydHub, set data_dir to "/input" and use the FloydHub data ID "R5KrjnANiKVhLWAkpXhNBe".


In [1]:
#I used a bunch of resources from the class such as course textbook, slack, forums, live support, and practice assignments. I also used the official tensorflow documentation.

#I ended up using an older tensorflow version 1.1 because while working on HW3 I ran into a lot of tensorflow version issues and incompatibilities on my machine. HW3 required tf1.0 and HW4 apparently required tf1.2. Luckly I, with the help of a Live Support TA, the issue was tracked down to version differences.. the day before my last day of class!!! W000t. Im just tired cause I just pulled two all nighters to finally finish these.


#I am glad I took this course. It made me actually sit down watch lectures, work through notebooks, an read all about the cutting edge of AI - Tensorflow.

#I want to thank the TAs who put up with and answered all my dumb questions as I figured this stuff out. Especially when they directed me to external, extra content to better understand the internals of what is happening in deep learning.


#PS. GANs are my new favorite thing.

In [2]:
#########
#
# Configure machine as discribed in project 1 (ubuntu, conda installed, etc)

 
# OH man this takes a long time to download. ...
#   ModuleNotFoundError: No module named 'PIL'

#  conda install pillow

# NOTE: This was writen as a project for a Udacity course and Udacity support was used, such as the Udacity forums, Udacity's Live Help function, the course material and the course's slack channel (especially the TA Hours :)  Tensorflow documentation was also heavily used

## Help resources such as the course website, cudacity course material, udacity forums, course slack channel, course lectures and practice assignments , and course live support were helpful in finishing this home work
# Of the course material amd practice assignments, the DCGAN and Intro_to_GANs assignements were particularly helpful.

#

In [3]:
data_dir = './data'

# FloydHub - Use with data ID "R5KrjnANiKVhLWAkpXhNBe"
#data_dir = '/input'


"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import helper

helper.download_extract('mnist', data_dir)
helper.download_extract('celeba', data_dir)


Found mnist Data
Found celeba Data

Explore the Data

MNIST

As you're aware, the MNIST dataset contains images of handwritten digits. You can view the first number of examples by changing show_n_images.


In [4]:
show_n_images = 25

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
%matplotlib inline
import os
from glob import glob
from matplotlib import pyplot

mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'mnist/*.jpg'))[:show_n_images], 28, 28, 'L')
pyplot.imshow(helper.images_square_grid(mnist_images, 'L'), cmap='gray')


Out[4]:
<matplotlib.image.AxesImage at 0x7f93711d2898>

CelebA

The CelebFaces Attributes Dataset (CelebA) dataset contains over 200,000 celebrity images with annotations. Since you're going to be generating faces, you won't need the annotations. You can view the first number of examples by changing show_n_images.


In [5]:
show_n_images = 25
show_n_images = 66

"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
mnist_images = helper.get_batch(glob(os.path.join(data_dir, 'img_align_celeba/*.jpg'))[:show_n_images], 28, 28, 'RGB')
pyplot.imshow(helper.images_square_grid(mnist_images, 'RGB'))


Out[5]:
<matplotlib.image.AxesImage at 0x7f936dff4a20>

Preprocess the Data

Since the project's main focus is on building the GANs, we'll preprocess the data for you. The values of the MNIST and CelebA dataset will be in the range of -0.5 to 0.5 of 28x28 dimensional images. The CelebA images will be cropped to remove parts of the image that don't include a face, then resized down to 28x28.

The MNIST images are black and white images with a single color channel while the CelebA images have 3 color channels (RGB color channel).

Build the Neural Network

You'll build the components necessary to build a GANs by implementing the following functions below:

  • model_inputs
  • discriminator
  • generator
  • model_loss
  • model_opt
  • train

Check the Version of TensorFlow and Access to GPU

This will check to make sure you have the correct version of TensorFlow and access to a GPU


In [6]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
from distutils.version import LooseVersion
import warnings
import tensorflow as tf

# Check TensorFlow Version
assert LooseVersion(tf.__version__) >= LooseVersion('1.0'), 'Please use TensorFlow version 1.0 or newer.  You are using {}'.format(tf.__version__)
print('TensorFlow Version: {}'.format(tf.__version__))

# Check for a GPU
if not tf.test.gpu_device_name():
    warnings.warn('No GPU found. Please use a GPU to train your neural network.')
else:
    print('Default GPU Device: {}'.format(tf.test.gpu_device_name()))


TensorFlow Version: 1.1.0
Default GPU Device: /gpu:0

Input

Implement the model_inputs function to create TF Placeholders for the Neural Network. It should create the following placeholders:

  • Real input images placeholder with rank 4 using image_width, image_height, and image_channels.
  • Z input placeholder with rank 2 using z_dim.
  • Learning rate placeholder with rank 0.

Return the placeholders in the following the tuple (tensor of real input images, tensor of z data)


In [7]:
import problem_unittests as tests

def model_inputs(image_width, image_height, image_channels, z_dim):
    """
    Create the model inputs
    :param image_width: The input image width
    :param image_height: The input image height
    :param image_channels: The number of image channels
    :param z_dim: The dimension of Z
    :return: Tuple of (tensor of real input images, tensor of z data, learning rate)
    """
    # done: Implement Function
    
    #realinputimages = tf.placeholder(dtype=tf.int32,shape=[None,None,None,None])
  
    #realinputimages = tf.placeholder(dtype=tf.int32,shape=[image_width,image_height,image_channels,None])
    #zinput = tf.placeholder(dtype=tf.int32,shape=[z_dim,None])
    #lr = tf.placeholder(dtype=tf.float32,shape=[])
    
    
    #AssertionError: Real Input has wrong shape.  Found [28, 28, 3, None]
    # looks like it needs to be the reverse direction for some reason...
    realinputimages = tf.placeholder(dtype=tf.float32,shape=[None,image_width,image_height,image_channels])
    zinput = tf.placeholder(dtype=tf.float32,shape=[None,z_dim])
    lr = tf.placeholder(dtype=tf.float32,shape=[])
    
    #ValueError: Tensor conversion requested dtype float32 for Tensor with dtype int32: 'Tensor("generator/batch_normalization/Reshape_2:0", shape=(512,), dtype=int32)'
    
    return realinputimages, zinput, lr


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_inputs(model_inputs)


Tests Passed

Discriminator

Implement discriminator to create a discriminator neural network that discriminates on images. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "discriminator" to allow the variables to be reused. The function should return a tuple of (tensor output of the discriminator, tensor logits of the discriminator).


In [8]:
# THis reminds me of the Intro_to_GANs project.
# reference from here on out: Intro_to_GANs_Solution. See : /deep-learning/gan_mnist/Intro_to_GANs_Solution.ipynb project, part of the course and related material

# aLSO the DCGANS tutorial / class material


def discriminator(images, reuse=False):
    """
    Create the discriminator network
    :param images: Tensor of input image(s)
    :param reuse: Boolean if the weights should be reused
    :return: Tuple of (tensor output of the discriminator, tensor logits of the discriminator)
    """
    # done: Implement Function
#https://www.tensorflow.org/api_docs/python/tf/variable_scope
    ## exxample: 
    #discriminator(x, n_units=128, reuse=False, alpha=0.01):
    #with tf.variable_scope("foo") as scope:
        #v = tf.get_variable("v", [1])
        #scope.reuse_variables()
        #v1 = tf.get_variable("v", [1])
    #assert v1 == v
    alpha=0.01
    n_units=128
   # with tf.variable_scope("discriminator") as scope:
   ##     if reuse:
   #         scope.reuse_variables()
       
    
   #     h1 = tf.layers.dense(images, n_units, activation=None)
   #    h1 = tf.maximum(alpha * h1, h1)
        
   #     logits = tf.layers.dense(h1, 1, activation=None)
   #     out = tf.sigmoid(logits)
        
        
        #ValueError: Discriminator Training(reuse=false) output has wrong rank.  Tensor Sigmoid:0 must have rank 2.  Received rank 4, shape (?, 28, 28, 1)
    #return out, logits
        

        ##
        #
        #      BASED ON THE DCGANs Material of The course
        #
        ##
#def generator(z, output_dim, reuse=False, alpha=0.2, training=True):
    #NameError: name 'x' is not defined
    with tf.variable_scope('discriminator', reuse=reuse):
        # Input layer is 32x32x3
        x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
        relu1 = tf.maximum(alpha * x1, x1)
        # 16x16x64
       
    
     #   x2 = tf.layers.conv2d(relu1, 128, 7, strides=2, padding='same')
       
        x2 = tf.layers.conv2d(relu1, 128, 5, strides=2, padding='same')
        bn2 = tf.layers.batch_normalization(x2, training=True)
        relu2 = tf.maximum(alpha * bn2, bn2)
        # 8x8x128
        
        
        
        #ValueError: Trying to share variable discriminator/conv2d/kernel, but specified shape (5, 5, 1, 64) and found shape (5, 5, 28, 64).
        x3 = tf.layers.conv2d(relu2, 256, 5, strides=2, padding='same')
        #x3 = tf.layers.conv2d(relu2, 256, 7, strides=2, padding='same')
        
        bn3 = tf.layers.batch_normalization(x3, training=True)
        relu3 = tf.maximum(alpha * bn3, bn3)
        # 4x4x256



        # Flatten it
        #flat = tf.reshape(relu3, (-1, 4*4*256))
    #flat = tf.reshape(relu3, (-1, 7*7*256))
       #
      #  flat = tf.reshape(x2, (-1, 7*7*256))
        
        flat = tf.reshape(relu3, (-1, 7*7*512))
        
        logits = tf.layers.dense(flat, 1)
        out = tf.sigmoid(logits)
        
        return out,logits
        

"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_discriminator(discriminator, tf)


Tests Passed

Generator

Implement generator to generate an image using z. This function should be able to reuse the variables in the neural network. Use tf.variable_scope with a scope name of "generator" to allow the variables to be reused. The function should return the generated 28 x 28 x out_channel_dim images.


In [9]:
## DCGANS was helpful in this cell and notebook


def generator(z, out_channel_dim, is_train=True):
    """
    Create the generator network
    :param z: Input z
    :param out_channel_dim: The number of channels in the output image
    :param is_train: Boolean if generator is being used for training
    :return: The tensor output of the generator
    """
    # TODO: Implement Function
  #  def generator(z, output_dim, reuse=False, alpha=0.2, training=True):
    #NameError: name 'reuse' is not defined
    #NameError: name 'alpha' is not defined
    alpha=0.2
    #(not is_train)
    is_ntrain = not is_train
    #with tf.variable_scope('generator', reuse=is_train:
    
    
    ## Playing with the numbers to get 28x28xoutdim ...
    
    
    with tf.variable_scope('generator', reuse=is_ntrain):
        # First fully connected layer
        #x1 = tf.layers.dense(z, 7*7*512)
        x1 = tf.layers.dense(z, 7*7*256)
        
        # Reshape it to start the convolutional stack
        #x1 = tf.reshape(x1, (-1, 7, 7, 512))
        x1 = tf.reshape(x1, (-1, 7, 7, 256))
        
        x1 = tf.layers.batch_normalization(x1, training=is_train)
        x1 = tf.maximum(alpha * x1, x1)
        
        
        #AssertionError: Generator output (is_train=True) has wrong shape.  Found [None, 32, 32, 5]
        
        x2 = tf.layers.conv2d_transpose(x1, 128, 5, strides=2, padding='same')
        x2 = tf.layers.batch_normalization(x2, training=is_train)
        x2 = tf.maximum(alpha * x2, x2)
 
        
        x3 = tf.layers.conv2d_transpose(x2, 128,5, strides=2, padding='same')
        x3 = tf.layers.batch_normalization(x3, training=is_train)
        x3 = tf.maximum(alpha * x3, x3)
        # 16x16x128 now
        
        # Output layer
        
#NameError: name 'output_dim' is not defined
        logits = tf.layers.conv2d_transpose(x3, out_channel_dim, 5, strides=1, padding='same')
        
        out = tf.tanh(logits)
        
    return out# ,logits
    #return None


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_generator(generator, tf)


Tests Passed

Loss

Implement model_loss to build the GANs for training and calculate the loss. The function should return a tuple of (discriminator loss, generator loss). Use the following functions you implemented:

  • discriminator(images, reuse=False)
  • generator(z, out_channel_dim, is_train=True)

In [10]:
def model_loss(input_real, input_z, out_channel_dim):
    """
    Get the loss for the discriminator and generator
    :param input_real: Images from the real dataset
    :param input_z: Z input
    :param out_channel_dim: The number of channels in the output image
    :return: A tuple of (discriminator loss, generator loss)
    """
    # done: Implement Function
    
    #NameError: name 'alpha' is not defined
    # Removing alpha from paramz
    
    g_model = generator(input_z, out_channel_dim)
    d_model_real, d_logits_real = discriminator(input_real)
    d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)

    
    """
    d_loss_real = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_model_real)))
    d_loss_fake = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_model_fake)))
    g_loss = tf.reduce_mean(
        tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_model_fake)))

   """
    
    #From help of Live Help TA
    d_loss_real = tf.reduce_mean( tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_real, labels=tf.ones_like(d_logits_real) * 0.9))

    d_loss_fake = tf.reduce_mean( tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.zeros_like(d_logits_fake)))

    g_loss = tf.reduce_mean( tf.nn.sigmoid_cross_entropy_with_logits(logits=d_logits_fake, labels=tf.ones_like(d_logits_fake)))

    
    d_loss = d_loss_real + d_loss_fake
    
    #return None, None # ValueError: None values not supported.
    return d_loss, g_loss


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_loss(model_loss)


Tests Passed

Optimization

Implement model_opt to create the optimization operations for the GANs. Use tf.trainable_variables to get all the trainable variables. Filter the variables with names that are in the discriminator and generator scope names. The function should return a tuple of (discriminator training operation, generator training operation).


In [11]:
def model_opt(d_loss, g_loss, learning_rate, beta1):
    """
    Get optimization operations
    :param d_loss: Discriminator loss Tensor
    :param g_loss: Generator loss Tensor
    :param learning_rate: Learning Rate Placeholder
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :return: A tuple of (discriminator training operation, generator training operation)
    """
    # done: Implement Function
    
    t_vars = tf.trainable_variables()
    d_vars = [var for var in t_vars if var.name.startswith('discriminator')]
    g_vars = [var for var in t_vars if var.name.startswith('generator')]

    #hELP from TA during Live Help
    update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS)
    gen_updates = [op for op in update_ops if op.name.startswith('generator')]
    with tf.control_dependencies(gen_updates):
        d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
        g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

  
    #with tf.control_dependencies(tf.get_collection(tf.GraphKeys.UPDATE_OPS)):
    #    d_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(d_loss, var_list=d_vars)
    #    g_train_opt = tf.train.AdamOptimizer(learning_rate, beta1=beta1).minimize(g_loss, var_list=g_vars)

    return d_train_opt, g_train_opt


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
tests.test_model_opt(model_opt, tf)


Tests Passed

Neural Network Training

Show Output

Use this function to show the current output of the generator during training. It will help you determine how well the GANs is training.


In [12]:
"""
DON'T MODIFY ANYTHING IN THIS CELL
"""
import numpy as np

def show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
    """
    Show example output for the generator
    :param sess: TensorFlow session
    :param n_images: Number of Images to display
    :param input_z: Input Z Tensor
    :param out_channel_dim: The number of channels in the output image
    :param image_mode: The mode to use for images ("RGB" or "L")
    """
    cmap = None if image_mode == 'RGB' else 'gray'
    z_dim = input_z.get_shape().as_list()[-1]
    example_z = np.random.uniform(-1, 1, size=[n_images, z_dim])

    samples = sess.run(
        generator(input_z, out_channel_dim, False),
        feed_dict={input_z: example_z})

    images_grid = helper.images_square_grid(samples, image_mode)
    pyplot.imshow(images_grid, cmap=cmap)
    pyplot.show()

Train

Implement train to build and train the GANs. Use the following functions you implemented:

  • model_inputs(image_width, image_height, image_channels, z_dim)
  • model_loss(input_real, input_z, out_channel_dim)
  • model_opt(d_loss, g_loss, learning_rate, beta1)

Use the show_generator_output to show generator output while you train. Running show_generator_output for every batch will drastically increase training time and increase the size of the notebook. It's recommended to print the generator output every 100 batches.


In [13]:
def train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode):
    """
    Train the GAN
    :param epoch_count: Number of epochs
    :param batch_size: Batch Size
    :param z_dim: Z dimension
    :param learning_rate: Learning Rate
    :param beta1: The exponential decay rate for the 1st moment in the optimizer
    :param get_batches: Function to get batches
    :param data_shape: Shape of the data
    :param data_image_mode: The image mode to use for images ("RGB" or "L")
    """
    # done: Build Model
    image_width = data_shape[0]
    image_height= data_shape[1]
    image_channels= data_shape[2]
    z_dim= data_shape[3]
    
    realinputimages, zinput, lr = model_inputs(image_width, image_height, image_channels, z_dim)
    d_loss, g_loss = model_loss(realinputimages, zinput, z_dim)
    model_opt  = model_opt(d_loss, g_loss, learning_rate, beta1)

    show_prgrezzGFX = 100
    show_prgrezzTXT = 10
    steps = 0
    
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for epoch_i in range(epoch_count):
            for batch_images in get_batches(batch_size):
                # done: Train Model
                
                
                steps += 1

                batch_z = np.random.uniform(-1, 1, size=(batch_size, z_size))

                # Run optimizers  
                
                
                ##### THESE ARE SOOO COOL> I" GOING TO USE OPTIMIZERS IN EVERYTHING FROM NOW ON
                ## also , ps., GANs are F* amazing
                # 
                _ = sess.run(net.d_opt, feed_dict={net.input_real: x, net.input_z: batch_z})
                _ = sess.run(net.g_opt, feed_dict={net.input_z: batch_z, net.input_real: x})

                if steps % show_prgrezzTXT == 0:
                    # At the end of each epoch, get the losses and print them out
                    train_loss_d = net.d_loss.eval({net.input_z: batch_z, net.input_real: x})
                    train_loss_g = net.g_loss.eval({net.input_z: batch_z})

                    print("Epoch {}/{}...".format(e+1, epochs),
                          "Discriminator Loss: {:.4f}...".format(train_loss_d),
                          "Generator Loss: {:.4f}".format(train_loss_g))
                    # Save losses to view after training
                    losses.append((train_loss_d, train_loss_g))

                if steps % show_prgrezzGFX == 0:
                # show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode):
                    show_generator_output(sess, n_images, input_z, out_channel_dim, image_mode)

MNIST

Test your GANs architecture on MNIST. After 2 epochs, the GANs should be able to generate images that look like handwritten digits. Make sure the loss of the generator is lower than the loss of the discriminator or close to 0.


In [14]:
batch_size = 64#None
z_dim = 100#None
learning_rate = 0.00025 #None
beta1 = 0.5# None

"""
#valuse used by DCGAN for comparison:
#####################
real_size = (32,32,3)
z_size = 100
learning_rate = 0.0002
batch_size = 128
epochs = 25
alpha = 0.2
beta1 = 0.5
"""
"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 2

mnist_dataset = helper.Dataset('mnist', glob(os.path.join(data_dir, 'mnist/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
          mnist_dataset.shape, mnist_dataset.image_mode)


---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-14-81f7424e76f8> in <module>()
     23 with tf.Graph().as_default():
     24     train(epochs, batch_size, z_dim, learning_rate, beta1, mnist_dataset.get_batches,
---> 25           mnist_dataset.shape, mnist_dataset.image_mode)

<ipython-input-13-20fb742c05c0> in train(epoch_count, batch_size, z_dim, learning_rate, beta1, get_batches, data_shape, data_image_mode)
     18 
     19     realinputimages, zinput, lr = model_inputs(image_width, image_height, image_channels, z_dim)
---> 20     d_loss, g_loss = model_loss(realinputimages, zinput, z_dim)
     21     model_opt  = model_opt(d_loss, g_loss, learning_rate, beta1)
     22 

<ipython-input-10-ba53c72d3303> in model_loss(input_real, input_z, out_channel_dim)
     17     g_model = generator(input_z, out_channel_dim)
     18     d_model_real, d_logits_real = discriminator(input_real)
---> 19     d_model_fake, d_logits_fake = discriminator(g_model, reuse=True)
     20 
     21 

<ipython-input-8-1ca59d405a4f> in discriminator(images, reuse)
     48     with tf.variable_scope('discriminator', reuse=reuse):
     49         # Input layer is 32x32x3
---> 50         x1 = tf.layers.conv2d(images, 64, 5, strides=2, padding='same')
     51         relu1 = tf.maximum(alpha * x1, x1)
     52         # 16x16x64

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/convolutional.py in conv2d(inputs, filters, kernel_size, strides, padding, data_format, dilation_rate, activation, use_bias, kernel_initializer, bias_initializer, kernel_regularizer, bias_regularizer, activity_regularizer, trainable, name, reuse)
    509       _reuse=reuse,
    510       _scope=name)
--> 511   return layer.apply(inputs)
    512 
    513 

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/base.py in apply(self, inputs, **kwargs)
    318       Output tensor(s).
    319     """
--> 320     return self.__call__(inputs, **kwargs)
    321 
    322 

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/base.py in __call__(self, inputs, **kwargs)
    284           input_shapes = [x.get_shape() for x in input_list]
    285           if len(input_shapes) == 1:
--> 286             self.build(input_shapes[0])
    287           else:
    288             self.build(input_shapes)

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/convolutional.py in build(self, input_shape)
    136                                   regularizer=self.kernel_regularizer,
    137                                   trainable=True,
--> 138                                   dtype=self.dtype)
    139     if self.use_bias:
    140       self.bias = vs.get_variable('bias',

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py in get_variable(name, shape, dtype, initializer, regularizer, trainable, collections, caching_device, partitioner, validate_shape, use_resource, custom_getter)
   1047       collections=collections, caching_device=caching_device,
   1048       partitioner=partitioner, validate_shape=validate_shape,
-> 1049       use_resource=use_resource, custom_getter=custom_getter)
   1050 get_variable_or_local_docstring = (
   1051     """%s

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py in get_variable(self, var_store, name, shape, dtype, initializer, regularizer, trainable, collections, caching_device, partitioner, validate_shape, use_resource, custom_getter)
    946           collections=collections, caching_device=caching_device,
    947           partitioner=partitioner, validate_shape=validate_shape,
--> 948           use_resource=use_resource, custom_getter=custom_getter)
    949 
    950   def _get_partitioned_variable(self,

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py in get_variable(self, name, shape, dtype, initializer, regularizer, reuse, trainable, collections, caching_device, partitioner, validate_shape, use_resource, custom_getter)
    347           reuse=reuse, trainable=trainable, collections=collections,
    348           caching_device=caching_device, partitioner=partitioner,
--> 349           validate_shape=validate_shape, use_resource=use_resource)
    350     else:
    351       return _true_getter(

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/base.py in variable_getter(getter, name, shape, dtype, initializer, regularizer, trainable, **kwargs)
    273           name, shape, initializer=initializer, regularizer=regularizer,
    274           dtype=dtype, trainable=trainable,
--> 275           variable_getter=functools.partial(getter, **kwargs))
    276 
    277     # Build (if necessary) and call the layer, inside a variable scope.

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/layers/base.py in _add_variable(self, name, shape, dtype, initializer, regularizer, trainable, variable_getter)
    226                                initializer=initializer,
    227                                dtype=dtype,
--> 228                                trainable=trainable and self.trainable)
    229     # TODO(sguada) fix name = variable.op.name
    230     if variable in existing_variables:

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py in _true_getter(name, shape, dtype, initializer, regularizer, reuse, trainable, collections, caching_device, partitioner, validate_shape, use_resource)
    339           trainable=trainable, collections=collections,
    340           caching_device=caching_device, validate_shape=validate_shape,
--> 341           use_resource=use_resource)
    342 
    343     if custom_getter is not None:

~/.conda/envs/dlndtf1.2/lib/python3.6/site-packages/tensorflow/python/ops/variable_scope.py in _get_single_variable(self, name, shape, dtype, initializer, regularizer, partition_info, reuse, trainable, collections, caching_device, validate_shape, use_resource)
    656         raise ValueError("Trying to share variable %s, but specified shape %s"
    657                          " and found shape %s." % (name, shape,
--> 658                                                    found_var.get_shape()))
    659       if not dtype.is_compatible_with(found_var.dtype):
    660         dtype_str = dtype.name

ValueError: Trying to share variable discriminator/conv2d/kernel, but specified shape (5, 5, 1, 64) and found shape (5, 5, 28, 64).

CelebA

Run your GANs on CelebA. It will take around 20 minutes on the average GPU to run one epoch. You can run the whole epoch or stop when it starts to generate realistic faces.


In [ ]:
batch_size = None
z_dim = None
learning_rate = None
beta1 = None


"""
DON'T MODIFY ANYTHING IN THIS CELL THAT IS BELOW THIS LINE
"""
epochs = 1

celeba_dataset = helper.Dataset('celeba', glob(os.path.join(data_dir, 'img_align_celeba/*.jpg')))
with tf.Graph().as_default():
    train(epochs, batch_size, z_dim, learning_rate, beta1, celeba_dataset.get_batches,
          celeba_dataset.shape, celeba_dataset.image_mode)

Submitting This Project

When submitting this project, make sure to run all the cells before saving the notebook. Save the notebook file as "dlnd_face_generation.ipynb" and save it as a HTML file under "File" -> "Download as". Include the "helper.py" and "problem_unittests.py" files in your submission.