Determine which page (i.e. item_page, search_results) the given URL is pointing at

Get the resource pool of the worker_cls

Get the resource pool of the worker_cls

Get the resource pool of the worker_cls

get order of seq lengths to make partitions balanced, this is used in balacing sum of seqlength across dp ranks and microbatches Paramete

Number of USD spent

returns the object type of a task param

Returns list of dicts with keys "action", "template_id", and "obj_ids"

r"""Make an iterator from the DataProto. This is built upon that TensorDict can be used as a normal Pytorch dataset. See https://pytorch.org/t

extract objects after "you see" and before "your task is to:"

Prepare the input for the encoder. Pass it through pad, embedding, and rnn.

Note that this operation is in-place

This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch, num_key_value_heads, seqlen, he

This is the equivalent of torch.repeat_interleave(x, dim=2, repeats=n_rep). The hidden states go from (batch, seqlen, num_key_value_heads, he

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Rotates half the hidden dims of the input.

Generate similarity scores for all pairs ``<text,text_pair>``. The inputs can be ``1 -> 1``, ``1 -> N`` or ``N -> N``. In the ``1 - N

Set ulysses sequence parallel process group.

Get one step response

verify the batch and save as ``acc`` tensor

broadcast tensor in tp shards across mp_group

broadcast tensor in tp shards across mp_group

broadcast tensor in tp shards across mp_group

broadcast tensor in tp shards across mp_group

Convert speech tokens to audio waveform Conversion Process: 1. Validate speech tokens and convert to tensor format

fetch tensor in tp shards

fetch tensor in tp shards

Generate model response using vLLM Generation Process: 1. Use vLLM's generate method with configured sampling parame

解释相关性结果

Load all required models for TTS processing Model Components: 1. vLLM: Large language model for text-to-speech token

Returns web request result wrapped in BeautifulSoup object Arguments: url (`str`): If no url or html is provided, use the cu