Module: tfm.utils

Stay organized with collections Save and categorize content based on your preferences.

Common TF utilities.


activations module: Activations package definition.


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)