Parses an expression with sympy.

Validate config options not implemented for Megatron backend

Validate config options not implemented for Megatron backend

Build the memory references. The memory buffers are built using the build_memory_buffer API. This API will allocate a weight buffer pointer to the

Check the consistency of the DataProto. Mainly for batch and non_tensor_batch We expose this function as a public one so that user can call th

Only collect the data from the tp=0 and pp=last and every dp ranks

treat pp as dp. Only collect data on tp=0

Computing reward given input_ids. The transformers should output a tensor with shape [batch_size, sequence_length], and the value at [EOS]

Convert Hydra configs and other special types to regular Python types.

Args: model_name: override_config_kwargs: Returns:

Create and initialize a trainer instance with the given config. Args: config: Configuration object with training parameters

name: name of the parameter param: training parameters infer_params (List[torch.Tensor]): a list of parameters all-gathered f

User passes in dp data. The data is dispatched to all tp/pp ranks with the same dp

treat pp as dp.

Extracts the install_requires list from setup.py using AST parsing.

for deepseek-ai/DeepSeek-R1-Distill-Qwen-7B

downsample the batch according to oversample_factor samples passing the filters will be prioritized

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

Patches flash attention forward to handle 3D position ids in mrope. (3, batch_size, seq_length)

Fold a batch dim from [bsz, xxx] into [new_bsz, bsz // new_bsz, xxx]

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

Gather the label from data. The value in label should be [0, vocab_size) Args: data: (..., vocab_size) label (torch.IntTensor) :

structure of qrecc passages.zip: - passages.zip - collection-paragraph - wayback - *.jsonl -

Generate ground truth response given a prompt.