Load data from the provided dataset_path or directly download the file(TODO).

r"""Load model and tokenizer for generator.

r""" Loads a single dataset and aligns it to the standard format.

r""" Aborts the processes recursively in a bottom-up way.

r""" Sets module as a leaf module to skip partitioning in deepspeed zero3.

r""" Aligned dataset: _prompt: [{"role": "user", "content": "..."}] * (2T - 1) _response: [{"role": "assistant", "content": "..."}

Wrap each block of the encoder to enable checkpointing.

Train actor model. Args: critic_model_group (PPORayActorGroup): critic model group. initial_model_group (PPORayActorG

r""" Casts projector output to half precision for fine-tuning quantized VLMs.

Building BM25 index based on bm25s library.

Building BM25 index based on Pyserini library. Reference: https://github.com/castorini/pyserini/blob/master/docs/usage-index.md#building-a-bm

Obtain the representation of documents based on the embedding model(BERT-based) and construct a faiss index.

r"""Constructing different indexes based on selective retrieval method.

r""" Checks the version of the required packages.

Compresses the given context, instruction and question. Args: context (List[str]): List of context strings that form the

Compute EM and F1 using your provided F1 function. Args: gold_list (list[str]): List of ground-truth answers. pred_list

Compute EM and F1 using your provided F1 function. Args: gold_list (list[str]): List of ground-truth answers. pred_list

r""" Computes loss for preference learning.

r""" Computes log probabilities of the reference model.

r""" Computes log probabilities of the reference model.

Concatenate the chosen and rejected inputs into a single tensor. Args: batch: A batch of data. Must contain the keys 'chosen_inpu

Concatenate the chosen and rejected inputs into a single tensor. Args: batch: A batch of data. Must contain the keys 'chosen_inpu

r""" Priority: PTQ-quantized (train/infer) > AutoGPTQ (export) > On-the-fly quantization (train/infer)

r""" Patches VLMs before loading them.

Recursively convert numpy objects in nested dictionaries or lists to native Python types.

将NumPy数组转换为嵌套列表结构