def testRank(self):
rank_op = lambda x: array_ops.rank_internal(x, optimize=False)
for dtype in self.numeric_types:
self._assertOpOutputMatchesExpected(
rank_op, dtype(7), expected=np.int32(0))
self._assertOpOutputMatchesExpected(
rank_op, np.array([[], []], dtype=dtype), expected=np.int32(2))
self._assertOpOutputMatchesExpected(
rank_op, np.array([-1, 1], dtype=dtype), expected=np.int32(1))
self._assertOpOutputMatchesExpected(
rank_op, np.array([[-1, 1]], dtype=dtype), expected=np.int32(2))
self._assertOpOutputMatchesExpected(
rank_op,
np.array([[-1], [1], [4]], dtype=dtype),
expected=np.int32(2))
def testRank(self):
rank_op = lambda x: array_ops.rank_internal(x, optimize=False)
for dtype in self.numeric_types:
self._assertOpOutputMatchesExpected(
rank_op, dtype(7), expected=np.int32(0))
self._assertOpOutputMatchesExpected(
rank_op, np.array([[], []], dtype=dtype), expected=np.int32(2))
self._assertOpOutputMatchesExpected(
rank_op, np.array([-1, 1], dtype=dtype), expected=np.int32(1))
self._assertOpOutputMatchesExpected(
rank_op, np.array([[-1, 1]], dtype=dtype), expected=np.int32(2))
self._assertOpOutputMatchesExpected(
rank_op,
np.array([[-1], [1], [4]], dtype=dtype),
expected=np.int32(2))
def testRank(self):
tf_val = tf.rank(tf.constant(0.0, shape=[1, 2, 3]))
c_val = tf.contrib.util.constant_value(tf_val)
self.assertEqual(np.ndarray, type(c_val))
self.assertEqual((), c_val.shape)
self.assertEqual(3, c_val)
# Repeat test using array_ops.rank_internal to avoid the optimization that
# happens in the rank function.
tf_val = array_ops.rank_internal(tf.constant(0.0, shape=[1, 2, 3]),
optimize=False)
c_val = tf.contrib.util.constant_value(tf_val)
self.assertEqual(np.ndarray, type(c_val))
self.assertEqual((), c_val.shape)
self.assertEqual(3, c_val)
self.assertEqual([3], c_val)
def testRank(self):
tf_val = tf.rank(tf.constant(0.0, shape=[1, 2, 3]))
c_val = tf.contrib.util.constant_value(tf_val)
self.assertEqual(np.ndarray, type(c_val))
self.assertEqual((), c_val.shape)
self.assertEqual(3, c_val)
# Repeat test using array_ops.rank_internal to avoid the optimization that
# happens in the rank function.
tf_val = array_ops.rank_internal(tf.constant(0.0, shape=[1, 2, 3]),
optimize=False)
c_val = tf.contrib.util.constant_value(tf_val)
self.assertEqual(np.ndarray, type(c_val))
self.assertEqual((), c_val.shape)
self.assertEqual(3, c_val)
self.assertEqual([3], c_val)
def assert_variables_initialized(var_list=None):
"""Returns an Op to check if variables are initialized.
NOTE: This function is obsolete and will be removed in 6 months. Please
change your implementation to use `report_uninitialized_variables()`.
When run, the returned Op will raise the exception `FailedPreconditionError`
if any of the variables has not yet been initialized.
Note: This function is implemented by trying to fetch the values of the
variables. If one of the variables is not initialized a message may be
logged by the C++ runtime. This is expected.
Args:
var_list: List of `Variable` objects to check. Defaults to the
value of `global_variables().`
Returns:
An Op, or None if there are no variables.
"""
if var_list is None:
var_list = global_variables() + local_variables()
# Backwards compatibility for old-style variables. TODO(touts): remove.
if not var_list:
var_list = []
for op in ops.get_default_graph().get_operations():
if op.type in ["Variable", "AutoReloadVariable"]:
var_list.append(op.outputs[0])
if not var_list:
return None
else:
ranks = []
for var in var_list:
with ops.colocate_with(var.op):
ranks.append(array_ops.rank_internal(var, optimize=False))
if len(ranks) == 1:
return ranks[0]
else:
return array_ops.pack(ranks)
def assert_variables_initialized(var_list=None):
"""Returns an Op to check if variables are initialized.
NOTE: This function is obsolete and will be removed in 6 months. Please
change your implementation to use `report_uninitialized_variables()`.
When run, the returned Op will raise the exception `FailedPreconditionError`
if any of the variables has not yet been initialized.
Note: This function is implemented by trying to fetch the values of the
variables. If one of the variables is not initialized a message may be
logged by the C++ runtime. This is expected.
Args:
var_list: List of `Variable` objects to check. Defaults to the
value of `all_variables().`
Returns:
An Op, or None if there are no variables.
"""
if var_list is None:
var_list = all_variables() + local_variables()
# Backwards compatibility for old-style variables. TODO(touts): remove.
if not var_list:
var_list = []
for op in ops.get_default_graph().get_operations():
if op.type in ["Variable", "AutoReloadVariable"]:
var_list.append(op.outputs[0])
if not var_list:
return None
else:
ranks = []
for var in var_list:
with ops.colocate_with(var.op):
ranks.append(array_ops.rank_internal(var, optimize=False))
if len(ranks) == 1:
return ranks[0]
else:
return array_ops.pack(ranks)
