A DataProto is a data structure that aims to provide a standard protocol for data exchange between functions. It contains a batch (TensorDi

Distributed interface.

Default DataLoader.

r"""A callback for saving the processor.

r"""General class for chat models. Backed by huggingface or vllm engines. Supports both sync and async methods. Sync methods: chat(), stream_

A container class used to store the whole dataset. Inside the class, each data sample will be stored in `Item` class. The properties of the datas

Data engine. Args: data_args: Data arguments.

r"""Arguments pertaining to which techniques we are going to fine-tuning with.

r"""Arguments pertaining to which model/config/tokenizer we are going to fine-tune or infer. The class on the most right will be displayed first.

r"""Arguments pertaining to what data we are going to input our model for training and evaluation.

A buffer to store samples for dynamic batch size.

Encoder class for encoding queries using a specified model. Attributes: model_name (str): The name of the model. model_

Sharding manager to support data resharding when using FSDP + Ulysses

Used to count mfu during training loop Example: flops_counter = FlopsCounter(config) flops_achieved, flops_promised = f

r"""Data collator for pairwise data.

Batching text queue for text data.

r"""Inherits Seq2SeqTrainer to compute generative metrics such as BLEU and ROUGE.

Distributed strategy.

A container class used to store and manipulate a sample within a dataset. Information related to this sample during training/inference will be st

Model loader. Args: model_args: Model arguments. is_trainable: Whether to train the model.

A stopping criteria that halts the text generation process if any specified stop word is encountered. Inspired by https://discuss.hugging

r"""Compute and store the average and current value.

ClipEncoder class for encoding queries using CLIP.

Plugin for data converters.

DataProtoFuture aims to eliminate actual data fetching on driver. By doing so, the driver doesn't have to wait for data so that asynchronou

Example for now.

r"""Dense retriever based on pre-built faiss index.

r"""A dummy optimizer used for the GaLore or APOLLO algorithm.

r"""A general engine to control the behaviors of Web UI.

Evaluator is used to summarize the results of all metrics.

A checkpoint manager that saves and loads - model - optimizer - lr_scheduler - extra_states in a SPMD way. We sa

r"""A callback for logging training and evaluation status.

A checkpoint manager that saves and loads - model - optimizer - lr_scheduler - extra_states in a SPMD way. We sa

A memory buffer is a contiguous torch tensor that may combine multiple tensors sharing with the underlying memory. It must have a unique ty

STEncoder class for encoding queries using SentenceTransformers. Attributes: model_name (str): The name of the model. m

Adaptive KL controller described in the paper: https://arxiv.org/pdf/1909.08593.pdf

r"""BM25 retriever based on pre-built pyserini index.

Wrapper replacing None outputs by empty tensors, which allows the use of checkpointing.

r"""Compute reward accuracy and support `batch_eval_metrics`.

r"""Compute accuracy and support `batch_eval_metrics`.

r"""Compute text similarity scores and support `batch_eval_metrics`. Wraps the tokenizer into metric functions, used in CustomSeq2SeqTrainer.

r"""Inherit PPOTrainer.

r"""Inherit Trainer for custom optimizer.

Default Data collator.

Plugin for adjusting dataset index.

Plugin for loading dataset.

Plugin for selecting dataset samples.

Fixed KL controller.