returnn.tensor.utils

Some helper utils.

returnn.tensor.utils.tensor_dict_fill_random_numpy_(tensor_dict: TensorDict, *, rnd: int | RandomState = 42, dyn_dim_max_sizes: Dict[Dim, int] | None = None, dyn_dim_min_sizes: Dict[Dim, int] | None = None)

Random fill with NumPy arrays.

Parameters:

• tensor_dict

• rnd

• dyn_dim_max_sizes – you can specify max sizes for dim tags with dynamic sizes. The fill random code makes sure that there is at least one entry where we reach the max size, so that the dim value will be the max size.

• dyn_dim_min_sizes

returnn.tensor.utils.tensor_fill_random_numpy_(x: Tensor, *, min_val: int = 0, max_val: int | None = None, rnd: RandomState, dyn_dim_max_sizes: Dict[Dim, int] | None = None, dyn_dim_min_sizes: Dict[Dim, int] | None = None) → bool

fill. return whether sth was filled

returnn.tensor.utils.tensor_dict_dims_random_seq_len_min_max(tensor_dict: TensorDict, seq_lens: None | int | Tuple[int, int] | Dict[str | Dim | None, int | Tuple[int, int]] = None) → Tuple[List[Dim], Dict[Dim, Tuple[int, int]]]

This is specifically intended to prepare the list of all dynamic dims from the tensor dict and the seq_len_min_max for get_random_seq_lens_for_dyn_dims().

Parameters:

• tensor_dict

• seq_lens – either fixed seq len, or take randint. per data key, or per dim, or same for all

Returns:

dims, seq_len_min_max

returnn.tensor.utils.get_random_seq_lens_for_dyn_dims(dims: Sequence[Dim], seq_len_min_max: Dict[Dim, Tuple[int, int]], *, batch_size: int = 1, rnd: int | RandomState = 1337) → Dict[Dim, ndarray]

Make random seq lens for dims.

Note that dim tags are not actually modified here, as we need to have this in a safe way, which might run in parallel to the main thread.

Parameters:

• dims – Note that the order matter, as we use complete_dyn_size() (or equivalent).

• seq_len_min_max

• batch_size

• rnd