Yield elements from data in chunks of batch_size.

Yield elements from data in chunks of batch_size, where each chunk size is a multiple of batch_size_multiple. This is an extended version of

Yield elements from data in chunks of batch_size, where each chunk size is a multiple of batch_size_multiple. This is an extended version of t

loss for discretized mixture logistic distribution Args: x: [batch, ] means: [batch, nmix] logscales: [batch, nmix]

topk for discretized mixture logistic distribution Args: means: [batch, nmix] logscales: [batch, nmix] logit_probs:,

Args: input: Tensor input tensor [batch, N1, N2, ..., Nl, in_features] mask: Tensor

Args: input: Tensor input tensor [batch, N1, N2, ..., Nl, in_features] mask: Tensor

Args: x: Tensor The random variable before flow h: list of object other conditional

Args: ctx: Tensor tensor [batch, features] src_mask: Tensor tensor for source mask [

Replace non-padding symbols with their position numbers. Position numbers begin at padding_idx+1. Padding symbols are ignored.

r"""Applies each `module` in :attr:`modules` in parallel on arguments contained in :attr:`inputs` (positional) and :attr:`kwargs_tup` (keyword)

Args: ctx: Tensor tensor [batch, features] src_mask: Tensor tensor for source mask [b

Args: src_sents: Tensor [batch, src_length] tensor for source sentences tgt_sents: Tensor [batch, tg

Args: src_sents: Tensor [batch, src_length] tensor for source sentences tgt_sents: Tensor [batch, tg

Args: src_sents: Tensor [batch, src_length] tensor for source sentences src_masks: Tensor [batch, sr

Args: model_dim: int the input dimension for keys, queries and values heads: int num