View source on GitHub |
Common TF utilities.
Modules
activations
module: Activations package definition.
Functions
assert_rank(...)
: Raises an exception if the tensor rank is not of the expected rank.
cross_replica_concat(...)
: Concatenates the given value
across (GPU/TPU) cores, along axis
.
get_activation(...)
: Maps an identifier to a Python function, e.g., "relu" => tf.nn.relu
.
get_replica_id(...)
: Gets replica id depending on the environment.
get_shape_list(...)
: Returns a list of the shape of tensor, preferring static dimensions.
is_special_none_tensor(...)
: Checks if a tensor is a special None Tensor.
pack_inputs(...)
: Pack a list of inputs
tensors to a tuple. (deprecated)
safe_mean(...)
: Computes a safe mean of the losses.
unpack_inputs(...)
: unpack a tuple of inputs
tensors to a tuple. (deprecated)