Trigger parameter synchronization after training step This ensures rollouter always uses the latest trained parameters

Unnormalize actions using dataset statistics

Validate the bounding box dimensions and aspect ratio.

Add critic worker to role mapping.

Simulate data stream

Add reference policy worker if KL loss or KL reward is used.

Add reward model worker if enabled.

Installs a new signal handler while preserving any existing handler. If an existing handler is present, it will be called _after_ the new han

Apply KL penalty to the token-level rewards. This function computes the KL divergence between the reference policy and current policy, then a

Reroute to the appropriate apply_rotary_pos_emb function depending on bshd (conventional) / thd (packed seq) format In Qwen2-VL, the sha

Assemble gen_batch_output from RolloutSample objects Assembles batches from RolloutSample objects, similar to the _post_generate_batch logic

Build the memory buffer given weight_buffer_meta Args: weight_buffer_meta: contains mapping from name to a dictionary containing shape an

Bulk offload group.

Calculate a value based on majority voting. This function identifies the most common value for a specified vote key in the data, then re

Calculates a score for the interaction, potentially considering aspects like partial exposure & in-context task switching. sh

Calls the specific API to perform retrieval with retry logic for various errors, using increasing delay between retries. Logs internal calls

Calls the remote sandbox API to execute code with retry logic for Gateway Timeout, using increasing delay between retries. Logs internal call

Calls the remote search API to perform retrieval with retry logic for various errors, using increasing delay between retries. Logs internal

线程安全：可被 FastAPI/uvicorn 并发调用。 Playwright 只在专用线程内执行。

截取高德地图截图，并返回图片byte数据

线程安全：可被 FastAPI/uvicorn 并发调用。 Playwright 只在专用线程内执行。

Returns the smallest integer greater than or equal to 'number' that is divisible by 'factor'.

Compute advantage for GRPO, operating only on Outcome reward (with only one scalar reward for each response). Args: token_level_

Compute the attention mask for the response part of the sequence. This function extracts the portion of the attention mask that corresponds to th

Compute the attention mask for the response part of the sequence. This function extracts the portion of the attention mask that corresponds to th

Compute rollout correction metrics from log probabilities during training. This function is used in the actor to compute metrics using the CURREN

This function returns a FusedWorker class. `FusedWorker.{class_name}` -> FusedClass Use `class_name` as a param to directly access t

Create Ray resource pools for distributed training. Initializes resource pools based on the resource pool specification, with each po

Create resource pool manager Args: config: Configuration object roles: List of roles that need to create resource pools

Create a dataset. Arguments: data_config: The data config. tokenizer (Tokenizer): The tokenizer. processor (Processor): T

Create a dataset.

name: name of the parameter train_params: training parameters infer_params (Iterable[torch.Tensor]): a iterator towards list of parameter

The keys of the environment variables that are used to configure the Worker.

Execute the tool. Args: instance_id: The instance id of the tool. parameters: The json string of the parameters of th

Execute a method on rank zero worker asynchronously. Args: method_name: Name of the method to execute *args: Position

Extract the equation from the solution string.

Extract tool calls from the responses. Args: responses_ids (List[int]): The ids of the responses. Returns: T

The training loop of PPO. The driver process only need to call the compute functions of the worker group through RPC to const

Returns the largest integer less than or equal to 'number' that is divisible by 'factor'.

Args: input_ids: input ids. shape (batch_size, seq_length) attention_mask: attention_mask. shape (batch_size, seq_le

Args: input_ids: input ids. shape (batch_size, seq_length) attention_mask: attention_mask. shape (batch_size, seq_le

Perform a forward pass and optionally a backward pass on a batch of data. Args: data: The input data for the forward pas

We assume: - The model takes input: (input_ids, attention_mask, position_ids). No rmpad for the input - The communication sha

r""" Copy paste LLaMa's forward https://github.com/linkedin/Liger-Kernel/blob/main/src/liger_kernel/transformers/model/llama.py This func

Generate text using the SGLang server asynchronously.

Generate sequence with token-in-token-out.

Generate tokens from prompt ids. Args: request_id (str): request id for sticky session. prompt_ids (List[int]): List

Generate tokens from prompt ids, used for async partial. Args: request_id (str): request id for sticky session. promp