Update the output dict and keep a key in self.output can be used as an attribute.

r""" Registers a chat template. To add the following chat template: ``` <s><user>user prompt here <model>model response here</s>

Append experience to replay buffer.

r""" Optionally checks the package version.

r""" Gets element by id. Example: top.lang, train.dataset

r""" Validates if this model accepts the input modalities.

Load dataset from config.

r""" Gets a list of responses of the chat model.

Automatically select generator class based on config.

r""" Forms a list of slots according to the inputs to encode.

r"""Automatically select retriever class based on config's retrieval method Args: config (dict): configuration with 'retrieval_method' ke

r""" Gets the path to saved model checkpoints.

r""" Loads pretrained tokenizer and optionally loads processor. Note: including inplace operation of model_args.

r""" Log `logs` on the various objects watching training, including stored metrics.

r""" Loads pretrained model.

r""" Processes visual inputs. Returns: (llava and paligemma) pixel_values: tensor with shape (B, C, H, W) Return

The evaluation process after finishing overall generation

Get judgement result. Args: item: dataset item, contains question, retrieval result... Returns: judgement: b

r""" Gets element by id. Example: top.lang, train.dataset

r""" Gets chat template and fixes the tokenizer.

Update the overall output field for each sample in the dataset.

r""" Pre-processes input messages before tokenization for VLMs.

r""" Adds elements to manager.

Compute the approximate KL divergence between two distributions. Schulman blog: http://joschu.net/blog/kl-approx.html Args: log_

r""" Returns a single pair of token ids representing prompt and response respectively.

r""" Adds elements to manager.