MCPcopy Create free account

hub / github.com/PRIME-RL/PRIME / functions

Functions2,604 in github.com/PRIME-RL/PRIME

↓ 1 callersFunction_exists
hdfs capable to check whether a file_path is exists
training/verl/utils/hdfs_io.py:43
↓ 1 callersFunction_expand_mask
Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`.
training/verl/utils/torch_functional.py:470
↓ 1 callersFunction_expand_mask
Expands attention_mask from `[bsz, seq_len]` to `[bsz, 1, tgt_seq_len, src_seq_len]`.
training/verl/models/llama/megatron/modeling_llama_megatron.py:59
↓ 1 callersMethod_expand_to_token_level
(self, data: DataProto, scores: torch.Tensor)
training/verl/workers/fsdp_workers.py:730
↓ 1 callersFunction_fix_a_slash_b
(string)
eval/utils/math_equivalence.py:34
↓ 1 callersFunction_fix_a_slash_b
(string)
eval/Math/Qwen25-Math/evaluation/parser.py:43
↓ 1 callersFunction_fix_a_slash_b
(string)
training/verl/utils/reward_score/evaluation_utils/math_util/math_normalize.py:56
↓ 1 callersFunction_fix_a_slash_b
(string)
data_preprocessing/math_util/math_normalize.py:56
↓ 1 callersFunction_fix_fracs
(string)
eval/utils/math_equivalence.py:3
↓ 1 callersFunction_fix_fracs
(string)
eval/Math/Qwen25-Math/evaluation/parser.py:11
↓ 1 callersFunction_fix_fracs
(string)
training/verl/utils/reward_score/evaluation_utils/math_util/math_normalize.py:24
↓ 1 callersFunction_fix_fracs
(string)
data_preprocessing/math_util/math_normalize.py:24
↓ 1 callersFunction_fix_sqrt
(string)
eval/utils/math_equivalence.py:57
↓ 1 callersFunction_fix_sqrt
(string)
eval/Math/Qwen25-Math/evaluation/parser.py:60
↓ 1 callersFunction_fix_sqrt
(string)
training/verl/utils/reward_score/evaluation_utils/math_util/math_normalize.py:81
↓ 1 callersFunction_fix_sqrt
(string)
data_preprocessing/math_util/math_normalize.py:81
↓ 1 callersMethod_forward_head
(self, hidden_states)
training/verl/models/llama/megatron/modeling_llama_megatron.py:302
↓ 1 callersMethod_forward_head
(self, hidden_states)
training/verl/models/llama/megatron/modeling_llama_megatron.py:559
↓ 1 callersMethod_forward_micro_batch
(self, micro_batch)
training/verl/workers/fsdp_workers.py:721
↓ 1 callersMethod_generate_minibatch
(self, prompts: DataProto)
training/verl/workers/rollout/hf_rollout.py:51
↓ 1 callersMethod_get_executor_cls
(cls, engine_config: EngineConfig)
training/verl/third_party/vllm/vllm_v_0_6_3/llm_engine_sp.py:355
↓ 1 callersMethod_get_executor_cls
(cls, engine_config: EngineConfig)
training/verl/third_party/vllm/vllm_v_0_5_4/llm_engine_sp.py:281
↓ 1 callersMethod_get_free_port
(self)
training/verl/single_controller/base/worker.py:58
↓ 1 callersFunction_get_model_architecture
(config: PretrainedConfig)
training/verl/third_party/vllm/vllm_v_0_3_1/model_loader.py:46
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_3_1/model_loader.py:123
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_6_3/dtensor_weight_loaders.py:371
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_6_3/megatron_weight_loaders.py:299
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_5_4/dtensor_weight_loaders.py:331
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_5_4/megatron_weight_loaders.py:298
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_4_2/dtensor_weight_loaders.py:260
↓ 1 callersFunction_get_model_weight_loader
(arch: str)
training/verl/third_party/vllm/vllm_v_0_4_2/megatron_weight_loaders.py:298
↓ 1 callersMethod_get_node_ip
(self)
training/verl/single_controller/base/worker.py:39
↓ 1 callersFunction_get_parallel_model_architecture_from_config
(config: PretrainedConfig, value=False)
training/verl/utils/model.py:243
↓ 1 callersMethod_get_weights_iterator
(self, actor_model: Union[PreTrainedModel, Dict])
training/verl/third_party/vllm/vllm_v_0_6_3/model_loader.py:199
↓ 1 callersMethod_get_weights_iterator
(self, actor_model: Union[PreTrainedModel, Dict])
training/verl/third_party/vllm/vllm_v_0_5_4/model_loader.py:181
↓ 1 callersMethod_get_weights_iterator
(self, actor_model: Union[PreTrainedModel, Dict])
training/verl/third_party/vllm/vllm_v_0_4_2/model_loader.py:170
↓ 1 callersMethod_init_cache_engine
(self)
training/verl/third_party/vllm/vllm_v_0_6_3/worker.py:235
↓ 1 callersMethod_init_cache_engine
(self)
training/verl/third_party/vllm/vllm_v_0_5_4/worker.py:229
↓ 1 callersMethod_init_cache_engine
(self)
training/verl/third_party/vllm/vllm_v_0_4_2/worker.py:199
↓ 1 callersMethod_init_cache_sp
Profiles the memory usage and initializes the KV cache.
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:176
↓ 1 callersFunction_init_distributed_environment
Initialize the distributed environment.
training/verl/third_party/vllm/vllm_v_0_3_1/worker.py:270
↓ 1 callersMethod_init_executor
(self, model, distributed_init_method)
training/verl/third_party/vllm/vllm_v_0_6_3/spmd_gpu_executor.py:74
↓ 1 callersMethod_init_executor
(self, model, distributed_init_method)
training/verl/third_party/vllm/vllm_v_0_5_4/spmd_gpu_executor.py:66
↓ 1 callersMethod_init_executor
(self, model, distributed_init_method)
training/verl/third_party/vllm/vllm_v_0_4_2/spmd_gpu_executor.py:63
↓ 1 callersMethod_init_head
(self)
training/verl/models/llama/megatron/modeling_llama_megatron.py:290
↓ 1 callersMethod_init_head
(self)
training/verl/models/llama/megatron/modeling_llama_megatron.py:547
↓ 1 callersMethod_init_rope
(self)
training/verl/models/llama/megatron/layers/parallel_attention.py:204
↓ 1 callersMethod_init_tokenizer
(self, tokenizer, **tokenizer_init_kwargs)
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:138
↓ 1 callersMethod_init_tokenizer
(self, tokenizer, **tokenizer_init_kwargs)
training/verl/third_party/vllm/vllm_v_0_6_3/llm_engine_sp.py:337
↓ 1 callersMethod_init_tokenizer
(self, tokenizer, **tokenizer_init_kwargs)
training/verl/third_party/vllm/vllm_v_0_5_4/llm_engine_sp.py:263
↓ 1 callersMethod_init_tokenizer
(self, tokenizer, **tokenizer_init_kwargs)
training/verl/third_party/vllm/vllm_v_0_4_2/llm_engine_sp.py:229
↓ 1 callersMethod_init_with_detached_workers
(self, worker_names)
training/verl/single_controller/ray/base.py:209
↓ 1 callersMethod_init_with_resource_pool
(self, resource_pool, ray_cls_with_init, bin_pack, detached)
training/verl/single_controller/ray/base.py:214
↓ 1 callersMethod_init_workers_sp
(self, model, distributed_init_method: str)
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:149
↓ 1 callersMethod_init_workers_sp
(self, model, distributed_init_method: str)
training/verl/third_party/vllm/vllm_v_0_6_3/spmd_gpu_executor.py:80
↓ 1 callersMethod_init_workers_sp
(self, model, distributed_init_method: str)
training/verl/third_party/vllm/vllm_v_0_5_4/spmd_gpu_executor.py:72
↓ 1 callersMethod_init_workers_sp
(self, model, distributed_init_method: str)
training/verl/third_party/vllm/vllm_v_0_4_2/spmd_gpu_executor.py:69
↓ 1 callersFunction_inject_implicit_mixed_number
Automatically make a mixed number evalable e.g. 7 3/4 => 7+3/4
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:88
↓ 1 callersFunction_inject_implicit_mixed_number
Automatically make a mixed number evalable e.g. 7 3/4 => 7+3/4
data_preprocessing/math_util/__init__.py:88
↓ 1 callersFunction_inner_call_method
(_method)
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/evaluation/testing_util.py:667
↓ 1 callersFunction_inner_call_method
(_method)
training/verl/utils/reward_score/evaluation_utils/code_util/testing_util.py:658
↓ 1 callersFunction_is_float
(num: str)
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:54
↓ 1 callersFunction_is_float
(num: str)
data_preprocessing/math_util/__init__.py:54
↓ 1 callersFunction_is_int
(x: float)
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:62
↓ 1 callersFunction_is_int
(x: float)
data_preprocessing/math_util/__init__.py:62
↓ 1 callersMethod_is_worker_alive
(self, worker)
training/verl/single_controller/base/worker_group.py:110
↓ 1 callersFunction_last_boxed_only_string
(string)
eval/Math/aime/evaluate_aime.py:67
↓ 1 callersFunction_last_boxed_only_string
(string)
eval/Math/amc/evaluate_amc.py:73
↓ 1 callersFunction_last_boxed_only_string
(string)
eval/Math/math/evaluate_math.py:64
↓ 1 callersFunction_last_boxed_only_string
(string)
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:304
↓ 1 callersFunction_last_boxed_only_string
(string)
data_preprocessing/math_util/__init__.py:310
↓ 1 callersMethod_load_params_to_cuda
(self, pp_rank, to_empty=False)
training/verl/workers/hybrid_engine/megatron_vllm.py:92
↓ 1 callersFunction_make_causal_mask
Make causal mask used for bi-directional self-attention.
training/verl/utils/torch_functional.py:457
↓ 1 callersFunction_make_causal_mask
Make causal mask used for bi-directional self-attention.
training/verl/models/llama/megatron/modeling_llama_megatron.py:46
↓ 1 callersMethod_make_minibatch_iterator
(self, data: DataProto)
training/verl/workers/critic/dp_critic.py:56
↓ 1 callersMethod_make_minibatch_iterator
Make minibatch iterator for updating the actor See PPO paper for details. https://arxiv.org/abs/1707.06347
training/verl/workers/actor/dp_actor.py:89
↓ 1 callersFunction_materialize_futures
(*args, **kwargs)
training/verl/single_controller/base/decorator.py:379
↓ 1 callersFunction_megatron_calc_layer_map
Calculate the mapping of global layer_idx to local layer_idx Returns: layer_map (Dict: int -> tuple(int, int, int)): mapping f
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:45
↓ 1 callersFunction_megatron_calc_layer_map
Calculate the mapping of global layer_idx to local layer_idx Returns: layer_map (Dict: int -> tuple(int, int, int)): mapping f
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:21
↓ 1 callersFunction_mkdir
hdfs mkdir
training/verl/utils/hdfs_io.py:75
↓ 1 callersMethod_normalize_config_bsz
(self)
training/verl/trainer/fsdp_sft_trainer.py:81
↓ 1 callersMethod_optimizer_step
(self)
training/verl/workers/critic/dp_critic.py:63
↓ 1 callersMethod_optimizer_step
(self)
training/verl/workers/actor/dp_prime.py:65
↓ 1 callersMethod_optimizer_step
(self)
training/verl/workers/actor/dp_actor.py:99
↓ 1 callersFunction_parse_latex
Attempts to parse latex to an expression sympy can read.
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:36
↓ 1 callersFunction_parse_latex
Attempts to parse latex to an expression sympy can read.
data_preprocessing/math_util/__init__.py:36
↓ 1 callersMethod_post_process_outputs
(self, outputs: List[RequestOutput])
training/verl/third_party/vllm/vllm_v_0_3_1/llm.py:250
↓ 1 callersMethod_post_process_outputs
(self, request_outputs: List[RequestOutput])
training/verl/third_party/vllm/vllm_v_0_6_3/llm.py:174
↓ 1 callersMethod_post_process_outputs
(self, request_outputs: List[RequestOutput])
training/verl/third_party/vllm/vllm_v_0_5_4/llm.py:214
↓ 1 callersMethod_post_process_outputs
(self, request_outputs: List[RequestOutput])
training/verl/third_party/vllm/vllm_v_0_4_2/llm.py:281
↓ 1 callersMethod_post_process_params
For each param, if it is a tp-splited param, we all-gather from micro_dp group.
training/verl/workers/hybrid_engine/megatron_vllm.py:318
↓ 1 callersFunction_pre_process_inputs
(pad_token_id, prompt_token_ids: torch.Tensor)
training/verl/workers/rollout/vllm_rollout/vllm_rollout.py:49
↓ 1 callersMethod_pre_process_inputs
(self, prompt_token_ids: torch.Tensor)
training/verl/third_party/vllm/vllm_v_0_3_1/llm.py:242
↓ 1 callersMethod_pre_process_inputs
(self, prompt_token_ids: torch.Tensor)
training/verl/third_party/vllm/vllm_v_0_4_2/llm.py:273
↓ 1 callersMethod_prepare_decoder_attention_mask
(self, attention_mask, input_shape, inputs_embeds)
training/verl/models/llama/megatron/modeling_llama_megatron.py:98
↓ 1 callersMethod_prepare_sample
( self, seq_group_metadata_list: List[SequenceGroupMetadata], prompt_lens: List[int],
training/verl/third_party/vllm/vllm_v_0_3_1/model_runner.py:109
↓ 1 callersMethod_process_model_outputs
(self, output: SamplerOutput, scheduler_outputs: SchedulerOutputs)
training/verl/third_party/vllm/vllm_v_0_3_1/llm_engine_sp.py:545
↓ 1 callersFunction_process_parameter_names
(name)
training/verl/third_party/vllm/vllm_v_0_6_3/dtensor_weight_loaders.py:326
↓ 1 callersFunction_process_parameter_names
(name)
training/verl/third_party/vllm/vllm_v_0_5_4/dtensor_weight_loaders.py:287
↓ 1 callersFunction_process_parameter_names
(name)
training/verl/third_party/vllm/vllm_v_0_4_2/dtensor_weight_loaders.py:218
← previousnext →501–600 of 2,604, ranked by callers