Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/PRIME-RL/PRIME
/ functions
Functions
2,604 in github.com/PRIME-RL/PRIME
⨍
Functions
2,604
◇
Types & classes
287
↳
Endpoints
32
↓ 2 callers
Function
normalize
(answer, pi)
training/verl/utils/reward_score/evaluation_utils/math_util/grader.py:108
↓ 2 callers
Function
normalize
(answer, pi)
data_preprocessing/math_util/grader.py:109
↓ 2 callers
Function
normalize_pp_vpp_params
Normalize the pp vpp params into a complete named parameters. This is useful when gather parameters from pp ranks and passed to a model with
training/verl/utils/model.py:181
↓ 2 callers
Function
numeric_equal
(prediction: float, reference: float)
eval/Math/Qwen25-Math/evaluation/grader.py:266
↓ 2 callers
Method
offload_params_to_cpu
(self)
training/verl/workers/reward_model/megatron/reward_model.py:262
↓ 2 callers
Method
ping
Check if the tokenizer group is alive.
training/verl/third_party/vllm/vllm_v_0_5_4/tokenizer.py:35
↓ 2 callers
Method
postfix_nofunc
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2495
↓ 2 callers
Function
postprocess_completion
(executor, completion)
eval/utils/python_interpreter.py:137
↓ 2 callers
Method
postprocess_data
(self, data: DataProto)
training/verl/workers/hybrid_engine/base.py:32
↓ 2 callers
Method
preprocess_data
(self, data: DataProto)
training/verl/workers/hybrid_engine/base.py:29
↓ 2 callers
Function
process_humaneval_test
(sample, problems, example_test=False, is_mbpp=False, language="python")
eval/utils/evaluation_leetcode.py:110
↓ 2 callers
Method
prompts_to_outputs
( self, prompts: list[str | list[dict[str, str]]] )
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/runner/base_runner.py:108
↓ 2 callers
Function
read_problems
(evalset_file: str = HUMAN_EVAL)
eval/utils/data.py:11
↓ 2 callers
Function
repeat_kv
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
training/verl/models/llama/megatron/layers/parallel_attention.py:131
↓ 2 callers
Function
rule2text
(ctx)
eval/Math/Qwen25-Math/evaluation/latex2sympy/latex2sympy2.py:732
↓ 2 callers
Method
run_batch
(self, prompts: list[str | list[dict[str, str]]])
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/runner/base_runner.py:67
↓ 2 callers
Function
sort_and_extract_save_results
(scenario: Scenario, save_results: list[dict])
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/runner/scenario_router.py:141
↓ 2 callers
Method
spawn
spawn to a dictionary of worker groups, each with a subset of method with prefix.
training/verl/single_controller/ray/base.py:292
↓ 2 callers
Function
split_tuple
Split the elements in a tuple/interval, while handling well-formatted commas in large numbers
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:221
↓ 2 callers
Function
split_tuple
Split the elements in a tuple/interval, while handling well-formatted commas in large numbers
data_preprocessing/math_util/__init__.py:221
↓ 2 callers
Function
strip_string
(string)
training/verl/utils/reward_score/math.py:165
↓ 2 callers
Function
stripped_string_compare
(s1, s2)
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/evaluation/testing_util.py:645
↓ 2 callers
Function
stripped_string_compare
(s1, s2)
training/verl/utils/reward_score/evaluation_utils/code_util/testing_util.py:636
↓ 2 callers
Method
sync_model_weights
(self, actor_weights: Dict[str, torch.Tensor])
training/verl/third_party/vllm/vllm_v_0_3_1/llm.py:271
↓ 2 callers
Function
text_to_trajectory
eval/Math/Qwen25-Math/evaluation/trajectory.py:13
↓ 2 callers
Method
truncate
(s, max_length=400)
eval/Math/Qwen25-Math/evaluation/python_executor.py:144
↓ 2 callers
Method
update_actor
(self, data: DataProto)
training/verl/workers/fsdp_workers.py:313
↓ 2 callers
Function
update_dtensor_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_6_3/dtensor_weight_loaders.py:379
↓ 2 callers
Function
update_dtensor_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_5_4/dtensor_weight_loaders.py:339
↓ 2 callers
Function
update_dtensor_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_4_2/dtensor_weight_loaders.py:268
↓ 2 callers
Function
update_hf_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_6_3/hf_weight_loader.py:22
↓ 2 callers
Function
update_hf_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_5_4/hf_weight_loader.py:25
↓ 2 callers
Function
update_hf_weight_loader
()
training/verl/third_party/vllm/vllm_v_0_4_2/hf_weight_loader.py:25
↓ 2 callers
Method
update_options
(self, options: Dict)
training/verl/single_controller/ray/base.py:139
↓ 2 callers
Method
verify_with_parallel_config
( self, parallel_config: "ParallelConfig", )
training/verl/third_party/vllm/vllm_v_0_3_1/config.py:163
↓ 2 callers
Method
wait
(self)
data_preprocessing/stage4_judge.py:26
↓ 2 callers
Function
write_jsonl_file
(file_path, data)
eval/Math/aime/evaluate_aime.py:37
↓ 2 callers
Function
write_jsonl_file
(file_path, data)
eval/Math/amc/evaluate_amc.py:37
↓ 1 callers
Method
ASSIGNMENT
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1557
↓ 1 callers
Method
BANG
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2528
↓ 1 callers
Method
CMD_CDOT
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2083
↓ 1 callers
Method
CMD_DIV
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2089
↓ 1 callers
Method
CMD_MOD
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2095
↓ 1 callers
Method
CMD_TIMES
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2080
↓ 1 callers
Method
COLON
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2092
↓ 1 callers
Method
DIV
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2086
↓ 1 callers
Method
EQUAL
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1560
↓ 1 callers
Method
EXP_E
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5334
↓ 1 callers
Method
FUNC_INT
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5289
↓ 1 callers
Method
FUNC_LIM
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5328
↓ 1 callers
Method
FUNC_PROD
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5322
↓ 1 callers
Method
FUNC_SQRT
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5307
↓ 1 callers
Method
FUNC_SUM
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5319
↓ 1 callers
Method
GT
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1569
↓ 1 callers
Method
GTE
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1572
↓ 1 callers
Method
IN
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1554
↓ 1 callers
Method
LTE
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1566
↓ 1 callers
Method
MATRIX_XRIGHTARROW
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1308
↓ 1 callers
Method
MUL
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2077
↓ 1 callers
Method
UNEQUAL
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1575
↓ 1 callers
Method
__init__
(self, stop_id_sequences)
eval/Math/Qwen25-Math/evaluation/model_utils.py:61
↓ 1 callers
Method
__init__
(self, numel: int, numel_padded: int, dtype: torch.dtype)
training/verl/utils/memory_buffer.py:30
↓ 1 callers
Method
__init__
(self, resource_pool: RayResourcePool, ray_cls_with_init: RayClassWithInitAr
training/verl/single_controller/ray/megatron.py:44
↓ 1 callers
Method
__init__
(self, config: LlamaConfig, megatron_config: ModelParallelConfig)
training/verl/models/llama/megatron/layers/parallel_decoder.py:35
↓ 1 callers
Method
__init__
(self, input_size, num_heads, num_key_value_heads,
training/verl/models/llama/megatron/layers/parallel_linear.py:23
↓ 1 callers
Method
__init__
( self, model: Union[nn.Module, Dict], # model itself or its parameter dict tokenizer:
training/verl/third_party/vllm/vllm_v_0_3_1/llm.py:87
↓ 1 callers
Method
__init__
( self, model: Union[nn.Module, Dict], # model itself or its parameter dict tokenizer:
training/verl/third_party/vllm/vllm_v_0_4_2/llm.py:89
↓ 1 callers
Method
_add_request
( self, prompt: Optional[str], sampling_params: SamplingParams, prompt_token_i
training/verl/third_party/vllm/vllm_v_0_3_1/llm.py:201
↓ 1 callers
Method
_add_request
( self, prompt: Optional[str], sampling_params: SamplingParams, prompt_token_i
training/verl/third_party/vllm/vllm_v_0_4_2/llm.py:232
↓ 1 callers
Method
_bind_worker_method
Bind the worker method to the WorkerGroup
training/verl/single_controller/base/worker_group.py:136
↓ 1 callers
Function
_bind_workers_method_to_parent
Binds the methods of each worker to the WorkerDict. Note that we only bind public methods that are decorated by register
training/verl/single_controller/ray/base.py:378
↓ 1 callers
Method
_block_until_all_workers_alive
(self)
training/verl/single_controller/base/worker_group.py:113
↓ 1 callers
Function
_broadcast_tp_shard_tensor_gate_up
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:214
↓ 1 callers
Function
_broadcast_tp_shard_tensor_gate_up
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:224
↓ 1 callers
Function
_broadcast_tp_shard_tensor_qkv
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:267
↓ 1 callers
Function
_broadcast_tp_shard_tensor_qkv
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:278
↓ 1 callers
Function
_broadcast_tp_shard_tensor_vocab
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:131
↓ 1 callers
Method
_build_critic_model_optimizer
(self, model_path, megatron_config
training/verl/workers/megatron_workers.py:446
↓ 1 callers
Method
_build_critic_model_optimizer
(self, config)
training/verl/workers/fsdp_workers.py:466
↓ 1 callers
Method
_build_dataloader
(self)
training/verl/trainer/fsdp_sft_trainer.py:92
↓ 1 callers
Method
_build_model
(self, config)
training/verl/workers/fsdp_workers.py:671
↓ 1 callers
Method
_build_model_optimizer
(self)
training/verl/trainer/fsdp_sft_trainer.py:135
↓ 1 callers
Method
_build_model_optimizer
(self, config, enable_gradient_checkpointing=False)
training/verl/workers/fsdp_workers.py:847
↓ 1 callers
Method
_build_param_buffer
Build the parameter buffer in each pp rank
training/verl/workers/hybrid_engine/megatron_vllm.py:82
↓ 1 callers
Method
_build_rm_model
(self, model_path, megatron_config: ModelParallelConfig, override_model_config)
training/verl/workers/megatron_workers.py:614
↓ 1 callers
Method
_build_rollout
(self)
training/verl/workers/megatron_workers.py:216
↓ 1 callers
Method
_build_rollout
(self)
training/verl/workers/fsdp_workers.py:224
↓ 1 callers
Method
_check_beam_search_early_stopping
( self, early_stopping: Union[bool, str], sampling_params: SamplingParams, bes
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:348
↓ 1 callers
Function
_check_dispatch_mode
(dispatch_mode)
training/verl/single_controller/base/decorator.py:366
↓ 1 callers
Function
_check_execute_mode
(execute_mode)
training/verl/single_controller/base/decorator.py:375
↓ 1 callers
Method
_check_stop
Stop the finished sequences.
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:681
↓ 1 callers
Function
_clean_numbers
Clean Numbers in the given string >>> _clean_numbers(None, "Hello 123") 'Hello 123' >>> _clean_numbers(None, "Hello 1234") 'Hell
eval/utils/util.py:69
↓ 1 callers
Method
_configure_before_init
(self, register_center_name: str, rank: int)
training/verl/single_controller/base/worker.py:102
↓ 1 callers
Method
_configure_with_meta
This function should only be called inside by WorkerGroup
training/verl/single_controller/base/worker.py:147
↓ 1 callers
Function
_copy
(from_path: str, to_path: str, timeout: int = None)
training/verl/utils/hdfs_io.py:113
↓ 1 callers
Method
_create_dataloader
(self)
training/verl/trainer/ppo/ray_trainer.py:254
↓ 1 callers
Method
_create_dataloader
(self)
training/verl/trainer/ppo/ray_trainer.py:757
↓ 1 callers
Method
_download
(self)
training/verl/utils/dataset/rm_dataset.py:70
↓ 1 callers
Method
_download
(self)
training/verl/utils/dataset/sft_dataset.py:69
↓ 1 callers
Method
_download
(self)
training/verl/utils/dataset/rl_dataset.py:95
← previous
next →
401–500 of 2,604, ranked by callers