def init_uninitialized_vars(target_vars: List[tf.Variable] = None) -> None:
"""Initialize all tf.Variables that have not already been initialized.
Equivalent to the following, but more efficient and does not bloat the tf graph:
tf.variables_initializer(tf.report_uninitialized_variables()).run()
"""
assert_tf_initialized()
if target_vars is None:
target_vars = tf.global_variables()
test_vars = []
test_ops = []
with tf.control_dependencies(
None): # ignore surrounding control_dependencies
for var in target_vars:
assert is_tf_expression(var)
try:
tf.get_default_graph().get_tensor_by_name(
var.name.replace(":0", "/IsVariableInitialized:0"))
except KeyError:
# Op does not exist => variable may be uninitialized.
test_vars.append(var)
with absolute_name_scope(var.name.split(":")[0]):
test_ops.append(tf.is_variable_initialized(var))
init_vars = [
var for var, inited in zip(test_vars, run(test_ops)) if not inited
]
run([var.initializer for var in init_vars])
def __init__(self, remote_vars, local_vars, *args, **kwargs):
if local_vars:
self.has_init = tf.is_variable_initialized(local_vars[0])
else:
self.has_init = tf.constant(True)
ready_for_local_init_op = tf.report_uninitialized_variables(
var_list=remote_vars)
self.init_op = tf.group(tf.initialize_variables(local_vars),
*tf.get_collection(tf.GraphKeys.LOCAL_INIT_OP))
if "scaffold" in kwargs:
# TODO(lmetz) I think this could technically be supported?
raise ValueError("Do not set scaffold on the session creator.")
scaffold = tf.train.Scaffold(
ready_for_local_init_op=ready_for_local_init_op,
ready_op=ready_for_local_init_op,
local_init_op=ready_for_local_init_op,
init_op=ready_for_local_init_op,
init_fn=self._maybe_initialize_local_vars_and_state_fn,
summary_op=tf.summary.merge([tf.summary.scalar("dummy", 0)]))
kwargs["scaffold"] = scaffold
super(WorkerSessionCreator, self).__init__(*args, **kwargs)
def _create_var(name: str, value_expr: TfExpression) -> TfExpression:
"""Internal helper for creating autosummary accumulators."""
assert not _finalized
name_id = name.replace("/", "_")
v = tf.cast(value_expr, _dtype)
if v.shape.is_fully_defined():
size = np.prod(v.shape.as_list())
size_expr = tf.constant(size, dtype=_dtype)
else:
size = None
size_expr = tf.reduce_prod(tf.cast(tf.shape(v), _dtype))
if size == 1:
if v.shape.ndims != 0:
v = tf.reshape(v, [])
v = [size_expr, v, tf.square(v)]
else:
v = [size_expr, tf.reduce_sum(v), tf.reduce_sum(tf.square(v))]
v = tf.cond(tf.is_finite(v[1]), lambda: tf.stack(v),
lambda: tf.zeros(3, dtype=_dtype))
with tfutil.absolute_name_scope("Autosummary/" +
name_id), tf.control_dependencies(None):
var = tf.Variable(tf.zeros(3, dtype=_dtype),
trainable=False) # [sum(1), sum(x), sum(x**2)]
update_op = tf.cond(tf.is_variable_initialized(var),
lambda: tf.assign_add(var, v),
lambda: tf.assign(var, v))
if name in _vars:
_vars[name].append(var)
else:
_vars[name] = [var]
return update_op
def get_uninitialized_variables(sess): global_vars = tf.global_variables() # print([str(i.name) for i in global_vars]) is_not_initialized = sess.run([tf.is_variable_initialized(var) for var in global_vars]) not_initialized_vars = [v for (v, f) in zip(global_vars, is_not_initialized) if not f] print([str(i.name) for i in not_initialized_vars]) return not_initialized_vars
def _initialize_uninitialized(self, sess):
global_vars = tf.global_variables()
is_not_initialized = sess.run(
[tf.is_variable_initialized(var) for var in global_vars])
not_initialized_vars = [
v for (v, f) in zip(global_vars, is_not_initialized) if not f
]
if not_initialized_vars:
sess.run(tf.variables_initializer(not_initialized_vars))
def initialize_uninitialized(sess):
global_vars = tf.global_variables()
is_not_initialized = sess.run([tf.is_variable_initialized(var) for var in global_vars])
not_initialized_vars = [v for (v, f) in zip(global_vars, is_not_initialized) if not f]
# for i in not_initialized_vars: # only for testing
# print(i.name)
if len(not_initialized_vars):
sess.run(tf.variables_initializer(not_initialized_vars))
return
def initialize_uninitialized_variables(sess):
"""
Only initialize the weights that have not yet been initialized by other
means, such as importing a metagraph and a checkpoint. It's useful when
extending an existing model.
"""
uninit_vars = []
uninit_tensors = []
for var in tf.global_variables():
uninit_vars.append(var)
uninit_tensors.append(tf.is_variable_initialized(var))
uninit_bools = sess.run(uninit_tensors)
uninit = zip(uninit_bools, uninit_vars)
uninit = [var for init, var in uninit if not init]
sess.run(tf.variables_initializer(uninit))
def initialize_uninitilized_global_variables(sess):
# from https://github.com/tensorflow/cleverhans/tree/master/cleverhans
# List all global variables
global_vars = tf.global_variables()
# Find initialized status for all variables
is_var_init = [tf.is_variable_initialized(var) for var in global_vars]
is_initialized = sess.run(is_var_init)
# List all variables that were not initialialized previously
not_initialized_vars = [
var for (var, init) in zip(global_vars, is_initialized) if not init
]
# Initialize all uninitialized variables found, if any
if len(not_initialized_vars):
sess.run(tf.variables_initializer(not_initialized_vars))
def init_uninited_vars(vars=None):
if vars is None: vars = tf.global_variables()
test_vars = []
test_ops = []
with tf.control_dependencies(
None): # ignore surrounding control_dependencies
for var in vars:
assert is_tf_expression(var)
try:
tf.get_default_graph().get_tensor_by_name(
var.name.replace(':0', '/IsVariableInitialized:0'))
except KeyError:
# Op does not exist => variable may be uninitialized.
test_vars.append(var)
with absolute_name_scope(var.name.split(':')[0]):
test_ops.append(tf.is_variable_initialized(var))
init_vars = [
var for var, inited in zip(test_vars, run(test_ops)) if not inited
]
run([var.initializer for var in init_vars])
def initialize_uninitialized_global_variables(sess):
"""
Only initializes the variables of a TensorFlow session that were not
already initialized.
:param sess: the TensorFlow session
:return:
"""
# List all global variables
global_vars = tf.global_variables()
# Find initialized status for all variables
is_var_init = [tf.is_variable_initialized(var) for var in global_vars]
is_initialized = sess.run(is_var_init)
# List all variables that were not initialized previously
not_initialized_vars = [var for (var, init) in
zip(global_vars, is_initialized) if not init]
# Initialize all uninitialized variables found, if any
if len(not_initialized_vars):
sess.run(tf.variables_initializer(not_initialized_vars))
def _create_autosummary_var(name, value_expr):
assert not _autosummary_finalized
v = tf.cast(value_expr, tf.float32)
if v.shape.ndims is 0:
v = [v, np.float32(1.0)]
elif v.shape.ndims is 1:
v = [tf.reduce_sum(v), tf.cast(tf.shape(v)[0], tf.float32)]
else:
v = [
tf.reduce_sum(v),
tf.reduce_prod(tf.cast(tf.shape(v), tf.float32))
]
v = tf.cond(tf.is_finite(v[0]), lambda: tf.stack(v), lambda: tf.zeros(2))
with tf.control_dependencies(None):
var = tf.Variable(tf.zeros(2)) # [numerator, denominator]
update_op = tf.cond(tf.is_variable_initialized(var),
lambda: tf.assign_add(var, v),
lambda: tf.assign(var, v))
if name in _autosummary_vars:
_autosummary_vars[name].append(var)
else:
_autosummary_vars[name] = [var]
return update_op
