:param input_ids: A tensor of shape `(batch_size, seq_len)`. :param input_embeddings: A tensor of shape `(batch_size, seq_len, d_mode

Load a tokenizer from a checkpoint.

load un-quantized pretrained model to cpu

load quantized model from local disk

- context_len allows for a rolling window context, allowing each prediction window to potentially condition on some context :param tok

- context_len allows for a rolling window context, allowing each prediction window to potentially condition on some context :param tok

Generate greedily until a stopping sequence :param requests: list A list of pairs (context, until) context: str

Generate greedily until a stopping sequence :param requests: list A list of pairs (context, until) context: str