def tiny(net, classes, num_anchors, training=False, center=True):
def batch_norm(net):
net = slim.batch_norm(net, center=center, scale=True, epsilon=1e-5, is_training=training)
if not center:
net = tf.nn.bias_add(net, slim.variable('biases', shape=[tf.shape(net)[-1]], initializer=tf.zeros_initializer()))
return net
scope = __name__.split('.')[-2] + '_' + inspect.stack()[0][3]
net = tf.identity(net, name='%s/input' % scope)
with slim.arg_scope([slim.layers.conv2d], kernel_size=[3, 3], weights_initializer=tf.truncated_normal_initializer(stddev=0.1), normalizer_fn=batch_norm, activation_fn=leaky_relu), slim.arg_scope([slim.layers.max_pool2d], kernel_size=[2, 2], padding='SAME'):
index = 0
channels = 16
for _ in range(5):
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.max_pool2d(net, scope='%s/max_pool%d' % (scope, index))
index += 1
channels *= 2
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.max_pool2d(net, stride=1, scope='%s/max_pool%d' % (scope, index))
index += 1
channels *= 2
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.conv2d(net, num_anchors * (5 + classes), kernel_size=[1, 1], activation_fn=None, scope='%s/conv' % scope)
net = tf.identity(net, name='%s/output' % scope)
return scope, net
python类variable()的实例源码
def darknet(net, classes, num_anchors, training=False, center=True):
def batch_norm(net):
net = slim.batch_norm(net, center=center, scale=True, epsilon=1e-5, is_training=training)
if not center:
net = tf.nn.bias_add(net, slim.variable('biases', shape=[tf.shape(net)[-1]], initializer=tf.zeros_initializer()))
return net
scope = __name__.split('.')[-2] + '_' + inspect.stack()[0][3]
net = tf.identity(net, name='%s/input' % scope)
with slim.arg_scope([slim.layers.conv2d], kernel_size=[3, 3], normalizer_fn=batch_norm, activation_fn=leaky_relu), slim.arg_scope([slim.layers.max_pool2d], kernel_size=[2, 2], padding='SAME'):
index = 0
channels = 32
for _ in range(2):
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.max_pool2d(net, scope='%s/max_pool%d' % (scope, index))
index += 1
channels *= 2
for _ in range(2):
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels / 2, kernel_size=[1, 1], scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.max_pool2d(net, scope='%s/max_pool%d' % (scope, index))
index += 1
channels *= 2
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels / 2, kernel_size=[1, 1], scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels / 2, kernel_size=[1, 1], scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
passthrough = tf.identity(net, name=scope + '/passthrough')
net = slim.layers.max_pool2d(net, scope='%s/max_pool%d' % (scope, index))
index += 1
channels *= 2
# downsampling finished
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels / 2, kernel_size=[1, 1], scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels / 2, kernel_size=[1, 1], scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
index += 1
with tf.name_scope(scope):
_net = reorg(passthrough)
net = tf.concat([_net, net], 3, name='%s/concat%d' % (scope, index))
net = slim.layers.conv2d(net, channels, scope='%s/conv%d' % (scope, index))
net = slim.layers.conv2d(net, num_anchors * (5 + classes), kernel_size=[1, 1], activation_fn=None, scope='%s/conv' % scope)
net = tf.identity(net, name='%s/output' % scope)
return scope, net
def _average_gradients(tower_grads, include_square=False):
"""Calculate the average gradient for each shared variable across all towers.
Note that this function provides a synchronization point across all towers.
Args:
tower_grads: List of lists of (gradient, variable) tuples. The outer list
is over individual gradients. The inner list is over the gradient
calculation for each tower.
Returns:
List of pairs of (gradient, variable) where the gradient has been averaged
across all towers.
"""
average_grads = []
average_grads_square = []
for grad_and_vars in zip(*tower_grads):
# Note that each grad_and_vars looks like the following:
# ((grad0_gpu0, var0_gpu0), ... , (grad0_gpuN, var0_gpuN))
grads = []
none_count = 0
for g, v in grad_and_vars:
if g == None:
none_count = none_count + 1
continue
# Add 0 dimension to the gradients to represent the tower.
expanded_g = tf.expand_dims(g, 0)
# Append on a 'tower' dimension which we will average over below.
grads.append(expanded_g)
if none_count==0:
# Average over the 'tower' dimension.
grad_cat = tf.concat(0, grads)
grad = tf.reduce_mean(grad_cat, 0)
# Keep in mind that the Variables are redundant because they are shared
# across towers. So .. we will just return the first tower's pointer to
# the Variable.
v = grad_and_vars[0][1]
grad_and_var = (grad, v)
average_grads.append(grad_and_var)
if include_square:
grad2 = tf.mul(grad_cat, grad_cat, name="square_gradient")
grad2 = tf.reduce_mean(grad2, 0)
average_grads_square.append((grad2, v))
elif none_count == len(grad_and_vars):
print("None gradient for %s" % (grad_and_vars[0][1].op.name))
else:
raise ValueError("None gradient error")
if include_square:
return average_grads, average_grads_square
else:
return average_grads
