chaserileyroberts

class TestClassifier():
  def test_classification(self):
    classifier = Classifier()
    dog_image = open_image('test/data/dog_image.png')
    classification = classifier.classify(dog_image)
    self.assertEquals(classification, "Dog")

chaserileyroberts

def make_convnet(input_image):
    net = slim.conv2d(input_image, 32, [11, 11], scope="conv1_11x11")
    net = slim.conv2d(input_image, 64, [5, 5], scope="conv2_5x5")
    net = slim.max_pool2d(net, [4, 4], stride=4, scope='pool1')
    net = slim.conv2d(input_image, 64, [5, 5], scope="conv3_5x5")
    net = slim.conv2d(input_image, 128, [3, 3], scope="conv4_3x3")
    net = slim.max_pool2d(net, [2, 2], scope='pool2')
    net = slim.conv2d(input_image, 128, [3, 3], scope="conv5_3x3")
    net = slim.max_pool2d(net, [2, 2], scope='pool3')
    net = slim.conv2d(input_image, 32, [1, 1], scope="conv6_1x1")

chaserileyroberts

def test_convnet():
  image = tf.placeholder(tf.float32, (None, 100, 100, 3)
  model = Model(image)
  sess = tf.Session()
  sess.run(tf.global_variables_initializer())
  before = sess.run(tf.trainable_variables())
  _ = sess.run(model.train, feed_dict={
               image: np.ones((1, 100, 100, 3)),
               })
  after = sess.run(tf.trainable_variables())

chaserileyroberts

  def make_convnet(image_input):
        # Try to normalize the input before convoluting
        net = slim.batch_norm(image_input)
        net = slim.conv2d(net, 32, [11, 11], scope="conv1_11x11")
        net = slim.conv2d(net, 64, [5, 5], scope="conv2_5x5")
        net = slim.max_pool2d(net, [4, 4], stride=4, scope='pool1')
        net = slim.conv2d(net, 64, [5, 5], scope="conv3_5x5")
        net = slim.conv2d(net, 128, [3, 3], scope="conv4_3x3")
        net = slim.max_pool2d(net, [2, 2], scope='pool2')
        net = slim.conv2d(net, 128, [3, 3], scope="conv5_3x3")

chaserileyroberts

class Model:
  def __init__(self, input, labels):
    """Classifier model
    Args:
      input: Input tensor of size (None, input_dims)
      label: Label tensor of size (None, 1). 
        Should be of type tf.int32.
    """
    prediction = self.make_network(input)
    # Prediction size is (None, 1).

chaserileyroberts

def test_loss():
  in_tensor = tf.placeholder(tf.float32, (None, 3))
  labels = tf.placeholder(tf.int32, None, 1))
  model = Model(in_tensor, labels)
  sess = tf.Session()
  loss = sess.run(model.loss, feed_dict={
    in_tensor:np.ones(1, 3),
    labels:[[1]]
  })
  assert loss != 0

chaserileyroberts

class GAN:
  def __init__(self, z_vector, true_images):
    # Pretend these are implemented.
    with tf.variable_scope("gen"):
      self.make_geneator(z_vector)
    with tf.variable_scope("des"):
      self.make_descriminator(true_images)
    opt = tf.AdamOptimizer()
    train_descrim = opt.minimize(self.descrim_loss)
    train_gen = opt.minimize(self.gen_loss)

chaserileyroberts

def test_gen_training():
  model = Model
  sess = tf.Session()
  gen_vars = tf.get_collection(tf.GraphKeys.VARIABLES, scope='gen')
  des_vars = tf.get_collection(tf.GraphKeys.VARIABLES, scope='des')
  before_gen = sess.run(gen_vars)
  before_des = sess.run(des_vars)
  # Train the generator.
  sess.run(model.train_gen)
  after_gen = sess.run(gen_vars)

chaserileyroberts

import mltest

def test_my_model():
  # Make placeholders for input into the model
  input_image = tf.placeholder(...)
  label = tf.placeholder(...)
  # Build your model
  model = Model(input_image, label)
  # Run the test suite!
  mltest.test_suite(

chaserileyroberts

import mltest
import pytest

def setup():
  mltest.setup()