Reduces a dictionary of metric lists by computing the mean of each list. Args: metrics: A dictionary mapping metric names to lists o

state transition function; here, takes in an action whose * logical and numerical preconds are fulfilled, and returns TRUE, * putting the result int

Dynamically update the active skill set during training.

broadcast tensor from rank0 across mp_group

broadcast tensor across mp_group

broadcast tensor from rank0 across mp_group

fetch tensor

fetch tensor

Create a base TransformerConfig with common parameters across different model architectures. TODO: (ycl) use dataclass or converter config?

Get a blank image as numpy array.

Run validation once. If use_skill is not None and val_envs supports it, set skill usage before running. Returns: (metric_dict

model: AutoModelForCausalLM

Deprecated. Please use copy_to_local instead.

exhaustive list of admissible commands

Get accurate font metrics for optimal layout calculation. Returns (average_char_width, line_height) Optimized for maximum density: m

Check if a path is a non-local (HDFS) path. Args: path (str): The path to check. Returns: bool: True if the path is an HDFS

Pad a DataProto to size divisible by size_divisor Args: size_divisor (int): size divisor Returns: data: (DataProto): the pad

Applies a monkey patch to the forward method of a given model class to enable Ulysses sequence parallelism input slicing.

Register a function with distributed execution configuration. This decorator registers a function with specific dispatch and execution modes

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

broadcast tensor across mp_group

broadcast tensor across mp_group

from https://github.com/sgl-project/sglang/blob/844e2f227ab0cce6ef818a719170ce37b9eb1e1b/python/sglang/srt/utils.py#L905 Broadcast inputs from sr

Returns a state representation built from various info sources.

Check and disable incompatible configurations for older Megatron version. Args: original_config (dict): The original model configura

Computes throughput metrics for PPO training. This function calculates performance metrics related to token processing speed, includ

Write all of the diagnostics from the current iteration level: int. (see logger.py docs) If the global logger level is higher than

Estimate the FLOPS based on the number of valid tokens in the current batch and the time taken. Args: batch_seqlens (Lis

Get the resource pool of the worker_cls

end of sentence token can be int or list: 1 or [1, 2] e.g. response_id = torch.tensor([[20, 10, 34, 1, 0, 0, 0],

Insert a rendered segment into the cache. Args: segment_text: The segment text (used as key) rendere