python类convolution2d()的实例源码

def residual_block(net, ch = 256, ch_inner = 128, scope = None, reuse = None, stride = 1):
        """
        Bottleneck v2
        """

        with slim.arg_scope([layers.convolution2d],
                            activation_fn = None,
                            normalizer_fn = None):
            with tf.variable_scope(scope, 'ResidualBlock', reuse = reuse):
                in_net = net

                if stride > 1:
                    net = layers.convolution2d(net, ch, kernel_size = 1, stride = stride)

                in_net = layers.batch_norm(in_net)
                in_net = tf.nn.relu(in_net)
                in_net = layers.convolution2d(in_net, ch_inner, 1)

                in_net = layers.batch_norm(in_net)
                in_net = tf.nn.relu(in_net)
                in_net = layers.convolution2d(in_net, ch_inner, 3, stride = stride)

                in_net = layers.batch_norm(in_net)
                in_net = tf.nn.relu(in_net)
                in_net = layers.convolution2d(in_net, ch, 1, activation_fn = None)

                net = tf.nn.relu(in_net + net)

        return net

def conv1d(self, net, num_ker, ker_size, stride):
        # 1D-convolution
        net = convolution2d(
            net,
            num_outputs=num_ker,
            kernel_size=[ker_size, 1],
            stride=[stride, 1],
            padding='SAME',
            activation_fn=None,
            normalizer_fn=None,
            weights_initializer=variance_scaling_initializer(),
            weights_regularizer=l2_regularizer(self.weight_decay),
            biases_initializer=tf.zeros_initializer)
        return net

def _cnn_to_mlp(convs, hiddens, dueling, inpt, num_actions, scope, reuse=False):
    with tf.variable_scope(scope, reuse=reuse):
        out = inpt
        with tf.variable_scope("convnet"):
            for num_outputs, kernel_size, stride in convs:
                out = layers.convolution2d(out,
                                           num_outputs=num_outputs,
                                           kernel_size=kernel_size,
                                           stride=stride,
                                           activation_fn=tf.nn.relu)
        out = layers.flatten(out)
        with tf.variable_scope("action_value"):
            action_out = out
            for hidden in hiddens:
                action_out = layers.fully_connected(action_out, num_outputs=hidden, activation_fn=tf.nn.relu)
            action_scores = layers.fully_connected(action_out, num_outputs=num_actions, activation_fn=None)

        if dueling:
            with tf.variable_scope("state_value"):
                state_out = out
                for hidden in hiddens:
                    state_out = layers.fully_connected(state_out, num_outputs=hidden, activation_fn=tf.nn.relu)
                state_score = layers.fully_connected(state_out, num_outputs=1, activation_fn=None)
            action_scores_mean = tf.reduce_mean(action_scores, 1)
            action_scores_centered = action_scores - tf.expand_dims(action_scores_mean, 1)
            return state_score + action_scores_centered
        else:
            return action_scores
        return out

def conv2d(self, net, num_ker, ker_size, stride):
        net = convolution2d(
            net,
            num_outputs=num_ker,
            kernel_size=[ker_size, ker_size],
            stride=[stride, stride],
            padding='SAME',
            activation_fn=None,
            normalizer_fn=None,
            weights_initializer=variance_scaling_initializer(),
            weights_regularizer=l2_regularizer(FLAGS.weight_decay),
            biases_initializer=tf.zeros_initializer)
        return net

def forward(image, num_actions):
    # Conv1
    out = layers.convolution2d(image, num_outputs=16, kernel_size=8, stride=4, activation_fn=tf.nn.relu, scope='conv1')
    out = layers.convolution2d(out, num_outputs=32, kernel_size=4, stride=2, activation_fn=tf.nn.relu, scope='conv2')
    out = layers.flatten(out, scope='flatten')
    out = layers.fully_connected(out, num_outputs=256, activation_fn=tf.nn.relu, scope='fc1')

    action_logprobs = tf.nn.log_softmax(layers.fully_connected(out, num_outputs=num_actions, activation_fn=None, scope='fc_actor'))
    value           = layers.fully_connected(out, num_outputs=1, activation_fn=None, scope='fc_critic')
    value           = tf.reshape(value, [-1])
    return action_logprobs, value

def dueling_model(img_in, num_actions, scope, noisy=False, reuse=False):
    """As described in https://arxiv.org/abs/1511.06581"""
    with tf.variable_scope(scope, reuse=reuse):
        out = img_in
        with tf.variable_scope("convnet"):
            # original architecture
            out = layers.convolution2d(out, num_outputs=32, kernel_size=8, stride=4, activation_fn=tf.nn.relu)
            out = layers.convolution2d(out, num_outputs=64, kernel_size=4, stride=2, activation_fn=tf.nn.relu)
            out = layers.convolution2d(out, num_outputs=64, kernel_size=3, stride=1, activation_fn=tf.nn.relu)
        out = layers.flatten(out)

        with tf.variable_scope("state_value"):
            if noisy:
                # Apply noisy network on fully connected layers
                # ref: https://arxiv.org/abs/1706.10295
                state_hidden = noisy_dense(out, name='noisy_fc1', size=512, activation_fn=tf.nn.relu)
                state_score = noisy_dense(state_hidden, name='noisy_fc2', size=1)
            else:
                state_hidden = layers.fully_connected(out, num_outputs=512, activation_fn=tf.nn.relu)
                state_score = layers.fully_connected(state_hidden, num_outputs=1, activation_fn=None)
        with tf.variable_scope("action_value"):
            if noisy:
                # Apply noisy network on fully connected layers
                # ref: https://arxiv.org/abs/1706.10295
                actions_hidden = noisy_dense(out, name='noisy_fc1', size=512, activation_fn=tf.nn.relu)
                action_scores = noisy_dense(actions_hidden, name='noisy_fc2', size=num_actions)
            else:
                actions_hidden = layers.fully_connected(out, num_outputs=512, activation_fn=tf.nn.relu)
                action_scores = layers.fully_connected(actions_hidden, num_outputs=num_actions, activation_fn=None)
            action_scores_mean = tf.reduce_mean(action_scores, 1)
            action_scores = action_scores - tf.expand_dims(action_scores_mean, 1)

        return state_score + action_scores

def policy_and_value_network(observations):
        # TODO: Baseline network, used in (Mnih et al., 2016)
        conv = tf_layers.convolution2d(observations, 16, 8, 4)
        conv = tf_layers.convolution2d(conv, 32, 4, 2)
        conv = tf_layers.flatten(conv)
        hidden_layer = tf_layers.fully_connected(conv, 128, activation_fn=tf.nn.relu)
        logits = tf_layers.linear(hidden_layer, env.actions)
        value = tf_layers.linear(hidden_layer, 1)
        # TODO: If you do not want to use baseline, uncomment the next line
#         value = tf.zeros([tf.shape(observations)[0], 1])
        return logits, value

def conv_learn(X, y, mode):
    # Ensure our images are 2d 
    X = tf.reshape(X, [-1, 36, 36, 1])
    # We'll need these in one-hot format
    y = tf.one_hot(tf.cast(y, tf.int32), 5, 1, 0)

    # conv layer will compute 4 kernels for each 5x5 patch
    with tf.variable_scope('conv_layer'):
        # 5x5 convolution, pad with zeros on edges
        h1 = layers.convolution2d(X, num_outputs=4,
                kernel_size=[5, 5], 
                activation_fn=tf.nn.relu)
        # 2x2 Max pooling, no padding on edges
        p1 = tf.nn.max_pool(h1, ksize=[1, 2, 2, 1],
                strides=[1, 2, 2, 1], padding='VALID')

        # Need to flatten conv output for use in dense layer
    p1_size = np.product(
              [s.value for s in p1.get_shape()[1:]])
    p1f = tf.reshape(p1, [-1, p1_size ])

    # densely connected layer with 32 neurons and dropout
    h_fc1 = layers.fully_connected(p1f,
             5,
             activation_fn=tf.nn.relu)
    drop = layers.dropout(h_fc1, keep_prob=0.5, is_training=mode == tf.contrib.learn.ModeKeys.TRAIN)

    logits = layers.fully_connected(drop, 5, activation_fn=None)
    loss = tf.losses.softmax_cross_entropy(y, logits)
    # Setup the training function manually
    train_op = layers.optimize_loss(
        loss,
        tf.contrib.framework.get_global_step(),
        optimizer='Adam',
        learning_rate=0.01)
    return tf.argmax(logits, 1), loss, train_op
# Use generic estimator with our function

def conv_learn(X, y, mode):
    # Ensure our images are 2d 
    X = tf.reshape(X, [-1, 36, 36, 1])
    # We'll need these in one-hot format
    y = tf.one_hot(tf.cast(y, tf.int32), 5, 1, 0)

    # conv layer will compute 4 kernels for each 5x5 patch
    with tf.variable_scope('conv_layer'):
        # 5x5 convolution, pad with zeros on edges
        h1 = layers.convolution2d(X, num_outputs=4,
                kernel_size=[5, 5], 
                activation_fn=tf.nn.relu)
        # 2x2 Max pooling, no padding on edges
        p1 = tf.nn.max_pool(h1, ksize=[1, 2, 2, 1],
                strides=[1, 2, 2, 1], padding='VALID')

        # Need to flatten conv output for use in dense layer
    p1_size = np.product(
              [s.value for s in p1.get_shape()[1:]])
    p1f = tf.reshape(p1, [-1, p1_size ])

    # densely connected layer with 32 neurons and dropout
    h_fc1 = layers.fully_connected(p1f,
             5,
             activation_fn=tf.nn.relu)
    drop = layers.dropout(h_fc1, keep_prob=0.5, is_training=mode == tf.contrib.learn.ModeKeys.TRAIN)

    logits = layers.fully_connected(drop, 5, activation_fn=None)
    loss = tf.losses.softmax_cross_entropy(y, logits)
    # Setup the training function manually
    train_op = layers.optimize_loss(
        loss,
        tf.contrib.framework.get_global_step(),
        optimizer='Adam',
        learning_rate=0.01)
    return tf.argmax(logits, 1), loss, train_op
# Use generic estimator with our function

def testInvalidDataFormat(self):
    height, width = 7, 9
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 3), seed=1)
      with self.assertRaisesRegexp(ValueError, 'data_format'):
        layers_lib.convolution2d(images, 32, 3, data_format='CHWN')

def testCreateConv(self):
    height, width = 7, 9
    with self.test_session():
      images = np.random.uniform(size=(5, height, width, 4))
      output = layers_lib.convolution2d(images, 32, [3, 3])
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, height, width, 32])
      weights = variables.get_variables_by_name('weights')[0]
      self.assertListEqual(weights.get_shape().as_list(), [3, 3, 4, 32])
      biases = variables.get_variables_by_name('biases')[0]
      self.assertListEqual(biases.get_shape().as_list(), [32])

def testCreateConvNCHW(self):
    height, width = 7, 9
    with self.test_session():
      images = np.random.uniform(size=(5, 4, height, width))
      output = layers_lib.convolution2d(images, 32, [3, 3], data_format='NCHW')
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, 32, height, width])
      weights = variables.get_variables_by_name('weights')[0]
      self.assertListEqual(weights.get_shape().as_list(), [3, 3, 4, 32])
      biases = variables.get_variables_by_name('biases')[0]
      self.assertListEqual(biases.get_shape().as_list(), [32])

def testCreateSquareConv(self):
    height, width = 7, 9
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 3), seed=1)
      output = layers_lib.convolution2d(images, 32, 3)
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, height, width, 32])

def testCreateConvWithTensorShape(self):
    height, width = 7, 9
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 3), seed=1)
      output = layers_lib.convolution2d(images, 32, images.get_shape()[1:3])
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, height, width, 32])

def testFullyConvWithCustomGetter(self):
    height, width = 7, 9
    with self.test_session():
      called = [0]

      def custom_getter(getter, *args, **kwargs):
        called[0] += 1
        return getter(*args, **kwargs)

      with variable_scope.variable_scope('test', custom_getter=custom_getter):
        images = random_ops.random_uniform((5, height, width, 32), seed=1)
        layers_lib.convolution2d(images, 64, images.get_shape()[1:3])
      self.assertEqual(called[0], 2)  # Custom getter called twice.

def testCreateVerticalConv(self):
    height, width = 7, 9
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 4), seed=1)
      output = layers_lib.convolution2d(images, 32, [3, 1])
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, height, width, 32])
      weights = variables.get_variables_by_name('weights')[0]
      self.assertListEqual(weights.get_shape().as_list(), [3, 1, 4, 32])
      biases = variables.get_variables_by_name('biases')[0]
      self.assertListEqual(biases.get_shape().as_list(), [32])

def testCreateHorizontalConv(self):
    height, width = 7, 9
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 4), seed=1)
      output = layers_lib.convolution2d(images, 32, [1, 3])
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(), [5, height, width, 32])
      weights = variables.get_variables_by_name('weights')[0]
      self.assertListEqual(weights.get_shape().as_list(), [1, 3, 4, 32])

def testCreateConvWithStride(self):
    height, width = 6, 8
    with self.test_session():
      images = random_ops.random_uniform((5, height, width, 3), seed=1)
      output = layers_lib.convolution2d(images, 32, [3, 3], stride=2)
      self.assertEqual(output.op.name, 'Conv/Relu')
      self.assertListEqual(output.get_shape().as_list(),
                           [5, height / 2, width / 2, 32])

def testCreateConvCreatesWeightsAndBiasesVars(self):
    height, width = 7, 9
    images = random_ops.random_uniform((5, height, width, 3), seed=1)
    with self.test_session():
      self.assertFalse(variables.get_variables('conv1/weights'))
      self.assertFalse(variables.get_variables('conv1/biases'))
      layers_lib.convolution2d(images, 32, [3, 3], scope='conv1')
      self.assertTrue(variables.get_variables('conv1/weights'))
      self.assertTrue(variables.get_variables('conv1/biases'))

def testCreateConvWithCollection(self):
    height, width = 7, 9
    images = random_ops.random_uniform((5, height, width, 3), seed=1)
    with ops.name_scope('fe'):
      conv = layers_lib.convolution2d(
          images, 32, [3, 3], outputs_collections='outputs', scope='Conv')
    output_collected = ops.get_collection('outputs')[0]
    self.assertEqual(output_collected.aliases, ['fe/Conv'])
    self.assertEqual(output_collected, conv)