def _gan28_discr(experiment=None, X=None, reuse=False, **kwargs):
lrelu = partial(leaky_relu, leakiness=experiment.leak)
h_dim = experiment.h_dim
logger = experiment.logger
msg = "D_SHAPE {} {} [reuse={}]"
logger.debug(msg.format("in", X.shape, reuse))
with tf.variable_scope("discriminator", reuse=reuse):
with tf.variable_scope("hidden1"):
d_hidden1 = conv2d(
X, filters=h_dim, kernel_size=5, strides=2, padding="same",
activation=lrelu, kernel_initializer=xavier_init())
logger.debug(msg.format("dh1", d_hidden1.shape, reuse))
with tf.variable_scope("hidden2"):
d_hidden2 = conv2d(
d_hidden1, filters=h_dim * 2, kernel_size=5, strides=2,
padding="same", activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh2", d_hidden2.shape, reuse))
with tf.variable_scope("hidden3"):
d_hidden2 = tf.reshape(
d_hidden2, [-1, np.prod(d_hidden2.shape[1:], dtype=int)])
d_hidden3 = dense(d_hidden2, 1024, activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh3", d_hidden3.shape, reuse))
with tf.variable_scope("output"):
d_out = dense(d_hidden3, 1, kernel_initializer=xavier_init())
logger.debug(msg.format("out", d_out.shape, reuse))
# define summaries on the last layer
d_summaries = stats_summaries(d_out)
return d_out, d_summaries
def _BuildGraph(self, x):
def _Quantize(x, r):
x = gen_array_ops.quantize_and_dequantize_v2(x, -r, r)
return x
def _DenseLayer(x, num_inputs, num_outputs, quantization_range, name):
"""Dense layer with quantized outputs.
Args:
x: input to the dense layer
num_inputs: number of input columns of x
num_outputs: number of output columns
quantization_range: the min/max range for quantization
name: name of the variable scope
Returns:
The output of the layer.
"""
with variable_scope.variable_scope(name):
kernel = variable_scope.get_variable(
'kernel',
shape=[num_inputs, num_outputs],
dtype=dtypes.float32,
initializer=keras.initializers.glorot_uniform())
bias = variable_scope.get_variable(
'bias',
shape=[num_outputs],
dtype=dtypes.float32,
initializer=keras.initializers.zeros())
x = math_ops.matmul(x, kernel)
x = _Quantize(x, quantization_range)
x = nn.bias_add(x, bias)
x = _Quantize(x, quantization_range)
return x
x = _Quantize(x, 1)
# Conv + Bias + Relu6
x = layers.conv2d(x, filters=32, kernel_size=3, use_bias=True)
x = nn.relu6(x)
# Conv + Bias + Relu6
x = layers.conv2d(x, filters=64, kernel_size=3, use_bias=True)
x = nn.relu6(x)
# Reduce
x = math_ops.reduce_mean(x, [1, 2])
x = _Quantize(x, 6)
# FC1
x = _DenseLayer(x, 64, 512, 6, name='dense')
x = nn.relu6(x)
# FC2
x = _DenseLayer(x, 512, 10, 25, name='dense_1')
x = array_ops.identity(x, name=OUTPUT_NODE_NAME)
return x
def _BuildGraph(self, x):
def _Quantize(x, r):
x = gen_array_ops.quantize_and_dequantize_v2(x, -r, r)
return x
def _DenseLayer(x, num_inputs, num_outputs, quantization_range, name):
"""Dense layer with quantized outputs.
Args:
x: input to the dense layer
num_inputs: number of input columns of x
num_outputs: number of output columns
quantization_range: the min/max range for quantization
name: name of the variable scope
Returns:
The output of the layer.
"""
with variable_scope.variable_scope(name):
kernel = variable_scope.get_variable(
'kernel',
shape=[num_inputs, num_outputs],
dtype=dtypes.float32,
initializer=init_ops.GlorotUniform())
bias = variable_scope.get_variable(
'bias',
shape=[num_outputs],
dtype=dtypes.float32,
initializer=init_ops.Zeros())
x = math_ops.matmul(x, kernel)
x = _Quantize(x, quantization_range)
x = nn.bias_add(x, bias)
x = _Quantize(x, quantization_range)
return x
x = _Quantize(x, 1)
# Conv + Bias + Relu6
x = layers.conv2d(x, filters=32, kernel_size=3, use_bias=True)
x = nn.relu6(x)
# Conv + Bias + Relu6
x = layers.conv2d(x, filters=64, kernel_size=3, use_bias=True)
x = nn.relu6(x)
# Reduce
x = math_ops.reduce_mean(x, [1, 2])
x = _Quantize(x, 6)
# FC1
x = _DenseLayer(x, 64, 512, 6, name='dense')
x = nn.relu6(x)
# FC2
x = _DenseLayer(x, 512, 10, 25, name='dense_1')
x = array_ops.identity(x, name=OUTPUT_NODE_NAME)
return x
def _can60_discr_32layer_auto(
experiment=None, X=None, reuse=False, use_constraints=None,
constraints_features=None, **kwargs):
lrelu = partial(leaky_relu, leakiness=experiment.leak)
h_dim = experiment.h_dim
logger = experiment.logger
msg = "D_SHAPE {} {} [reuse={}]"
logger.debug(msg.format("in", X.shape, reuse))
with tf.variable_scope("discriminator", reuse=reuse):
with tf.variable_scope("hidden1"):
d_hidden1 = conv2d(
X, filters=h_dim, kernel_size=5, strides=2, padding="same",
activation=lrelu, kernel_initializer=xavier_init())
logger.debug(msg.format("dh1", d_hidden1.shape, reuse))
with tf.variable_scope("hidden15"):
d_hidden15 = conv2d(
d_hidden1, filters=h_dim, kernel_size=5, strides=2,
padding="same", activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh15", d_hidden15.shape, reuse))
with tf.variable_scope("hidden2"):
d_hidden2 = conv2d(
d_hidden15, filters=h_dim * 2, kernel_size=5, strides=2,
padding="same", activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh2", d_hidden2.shape, reuse))
d_hidden2 = tf.reshape(
d_hidden2, [-1, np.prod(d_hidden2.shape[1:], dtype=int)])
logger.debug(msg.format("dh2", d_hidden2.shape, reuse))
with tf.variable_scope("hidden3"):
d_hidden3 = dense(d_hidden2, 1024, activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh3", d_hidden3.shape, reuse))
with tf.variable_scope("hidden4"):
d_hidden4 = dense(d_hidden3, 32, activation=lrelu,
kernel_initializer=xavier_init())
logger.debug(msg.format("dh4", d_hidden4.shape, reuse))
d_summaries_dh4 = stats_summaries(d_hidden4, "dh4_pre_cond")
with tf.variable_scope("shared_weights"):
d_out_kernel = tf.get_variable(
"d_out_kernel", shape=[32, 1], initializer=xavier_init())
logger.debug(msg.format("d_out_kernel", d_out_kernel.shape, reuse))
d_out_bias = tf.get_variable(
"d_out_bias", shape=[1, 1], initializer=xavier_init())
logger.debug(msg.format("d_out_bias", d_out_bias.shape, reuse))
def skip_constraints():
with tf.variable_scope("output"):
d_out = tf.add(tf.matmul(d_hidden4, d_out_kernel),
d_out_bias, name="d_out_{}".format(reuse))
logger.debug(msg.format("out", d_out.shape, reuse))
# define summaries on the last layer
d_summaries = tf.summary.merge(
[stats_summaries(d_out), d_summaries_dh4])
return d_out, d_summaries
def apply_constraints():
logger.debug("Using constraints: {}".format(
str(experiment.constraints)))
with tf.variable_scope("constrained_out"):
d_constraints_kernel = tf.get_variable(
"d_constraints_kernel",
shape=[constraints_features.shape[1], 1],
initializer=xavier_init())
logger.debug(msg.format(
"d_constraints_kernel", d_constraints_kernel.shape, reuse))
input_concat = tf.concat(
[d_hidden4, constraints_features],
axis=1, name="input_concat_{}".format(reuse))
logger.debug(msg.format(
"input_concat", input_concat.shape, reuse))
weight_concat = tf.concat(
[d_out_kernel, d_constraints_kernel],
axis=0, name="weight_concat_{}".format(reuse))
logger.debug(msg.format(
"weight_concat", weight_concat.shape, reuse))
d_constrained_out = tf.add(
tf.matmul(input_concat, weight_concat), d_out_bias,
name="d_constrained_out_{}".format(reuse))
logger.debug(msg.format(
"constrained_out", d_constrained_out.shape, reuse))
# define summaries on the last layer
d_summaries = tf.summary.merge(
[stats_summaries(d_constrained_out),
d_summaries_dh4])
return d_constrained_out, d_summaries
return tf.cond(tf.cast(use_constraints, tf.bool),
skip_constraints, apply_constraints)
