Modules
nest:
Functions for working with arbitrarily nested sequences of elements.
Classes
CriticalSection:
Critical section.
VariableDeviceChooser:
Device chooser for variables.
Functions
add_arg_scope(…):
Decorates a function with args so it can be used within an arg_scope.
add_model_variable(…):
Adds a variable to the GraphKeys.MODEL_VARIABLES collection.
arg_scope(…):
Stores the default arguments for the given set of list_ops.
arg_scoped_arguments(…):
Returns the list kwargs that arg_scope can set for a func.
assert_global_step(…):
DEPRECATED FUNCTION
assert_or_get_global_step(…):
Verifies that a global step tensor is valid or gets one if None is given.
assert_same_float_dtype(…):
Validate and return float type based on tensors and dtype.
assert_scalar(…):
assert_scalar_int(…):
Assert tensor is 0-D, of type tf.int32 or tf.int64.
assign_from_checkpoint(…):
Creates an operation to assign specific variables from a checkpoint.
assign_from_checkpoint_fn(…):
Returns a function that assigns specific variables from a checkpoint.
assign_from_values(…):
Creates an assignment operation from a given mapping.
assign_from_values_fn(…):
Returns a function that assigns specific variables from the given values.
convert_to_tensor_or_sparse_tensor(…):
Converts value to a SparseTensor or Tensor.
create_global_step(…):
Create global step tensor in graph. (deprecated)
current_arg_scope(…):
deprecated(…):
Decorator for marking functions or methods deprecated.
deprecated_arg_values(…):
Decorator for marking specific function argument values as deprecated.
deprecated_args(…):
Decorator for marking specific function arguments as deprecated.
filter_variables(…):
Filter a list of variables using regular expressions.
fuse_op(…):
Fuse subgraph between input_nodes and output_nodes into a single custom op.
get_global_step(…):
DEPRECATED FUNCTION
get_graph_from_inputs(…):
Returns the appropriate graph to use for the given inputs.
get_local_variables(…):
Gets the list of local variables, filtered by scope and/or suffix.
get_model_variables(…):
Gets the list of model variables, filtered by scope and/or suffix.
get_name_scope(…):
Returns the current name scope of the default graph.
get_or_create_global_step(…):
Returns and create (if necessary) the global step tensor. (deprecated)
get_trainable_variables(…):
Gets the list of trainable variables, filtered by scope and/or suffix.
get_unique_variable(…):
Gets the variable uniquely identified by that var_op_name.
get_variable_full_name(…):
Returns the full name of a variable.
get_variables(…):
Gets the list of variables, filtered by scope and/or suffix.
get_variables_by_name(…):
Gets the list of variables that were given that name.
get_variables_by_suffix(…):
Gets the list of variables that end with the given suffix.
get_variables_to_restore(…):
Gets the list of the variables to restore.
global_variable(…):
Create a variable with a value and add it to GraphKeys.GLOBAL_VARIABLES.
has_arg_scope(…):
Checks whether a func has been decorated with @add_arg_scope or not.
init_from_checkpoint(…):
Using assignment map initializes current variables with loaded tensors.
is_tensor(…):
Check whether x is of tensor type.
list_variables(…):
Returns list of all variables in the latest checkpoint.
load_and_remap_matrix_initializer(…):
Returns a var initializer for loading and remapping a 2-D (matrix) tensor.
load_checkpoint(…):
Returns CheckpointReader for latest checkpoint.
load_embedding_initializer(…):
Returns a variable initializer for loading pre-trained embeddings.
load_linear_multiclass_bias_initializer(…):
Loads pre-trained multi-class biases for linear models from checkpoint.
load_variable(…):
Returns a Tensor with the contents of the given variable in the checkpoint.
load_variable_slot_initializer(…):
Loads pre-trained multi-class slots for linear models from checkpoint.
local_variable(…):
Create a variable with a value and add it to GraphKeys.LOCAL_VARIABLES.
model_variable(…):
Gets an existing model variable with these parameters or creates a new one.
prepend_name_scope(…):
Prepends name scope to a name.
py_func(…):
Wraps a python function and uses it as a TensorFlow op.
reduce_sum_n(…):
Reduce tensors to a scalar sum.
remove_squeezable_dimensions(…):
Squeeze last dim if ranks of predictions and labels differ by 1. (deprecated)
sort(…):
Sorts a tensor.
strip_name_scope(…):
Removes name scope from a name.
variable(…):
Gets an existing variable with these parameters or creates a new one.
with_same_shape(…):
Assert tensors are the same shape, from the same graph.
with_shape(…):
Asserts tensor has expected shape.
zero_initializer(…):
Initialize ‘ref’ with all zeros, ref tensor should be uninitialized.