MCPcopy Create free account

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

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

⨍ Functions 2,604◇ Types & classes 287↳ Endpoints 32
↓ 5 callersFunction_run_cmd
(cmd: str, timeout=None)
training/verl/utils/hdfs_io.py:135
↓ 5 callersFunctionadd_rm_stat
(subset, type, d)
data_preprocessing/stage1_filter.py:20
↓ 5 callersFunctionadd_rm_stat
(subset, type, d)
data_preprocessing/stage3_merge.py:35
↓ 5 callersFunctionbroadcast_dict_tensor
TODO: optimize this. Technically, we only need one broadcast
training/verl/utils/torch_functional.py:160
↓ 5 callersFunctioncompute_position_id_with_mask
(mask)
training/verl/utils/model.py:177
↓ 5 callersMethodelementary_transform
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1192
↓ 5 callersMethodexecute_all_async
(self, method_name: str, *args, **kwargs)
training/verl/single_controller/ray/base.py:335
↓ 5 callersMethodgenerate_sequences
Generate sequences
training/verl/workers/rollout/base.py:35
↓ 5 callersFunctionget_init_weight_context_manager
(use_meta_tensor=True)
training/verl/utils/fsdp_utils.py:28
↓ 5 callersFunctionget_micro_data_parallel_world_size
()
training/verl/workers/hybrid_engine/megatron_vllm.py:423
↓ 5 callersFunctionget_model
Build the model.
training/verl/utils/megatron_utils.py:34
↓ 5 callersMethodinsert_output_evaluation
( self, output_list: list[str], code_list: list[str], graded_list: list[bool] )
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/benchmarks/code_execution.py:43
↓ 5 callersMethodmake_iterator
Make an iterator from the DataProto. This is built upon that TensorDict can be used as a normal Pytorch dataset. See https://pytorch.org/tenso
training/verl/protocol.py:349
↓ 5 callersFunctionmatch_answer
(response)
data_preprocessing/math_util/__init__.py:339
↓ 5 callersMethodmatrix_row
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1510
↓ 5 callersMethodmp_nofunc
(self, _p=0)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2204
↓ 5 callersFunctionmul_flat
(lh, rh)
eval/Math/Qwen25-Math/evaluation/latex2sympy/latex2sympy2.py:330
↓ 5 callersMethodslice
(self, index)
training/verl/protocol.py:122
↓ 5 callersMethodtransform_assignment
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1148
↓ 5 callersFunctionunwrap_model
(model, module_instances=ALL_MODULE_WRAPPER_CLASSNAMES)
training/verl/utils/megatron_utils.py:122
↓ 5 callersFunctionupdate_model_config
(module_config, override_config_kwargs)
training/verl/utils/model.py:42
↓ 4 callersMethodDIFFERENTIAL
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4384
↓ 4 callersMethodGREEK_CMD
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4278
↓ 4 callersMethodOTHER_SYMBOL_CMD
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4281
↓ 4 callersMethod__init__
(self, load_config: LoadConfig)
training/verl/third_party/vllm/vllm_v_0_6_3/model_loader.py:193
↓ 4 callersMethod__init__
(self, load_config: LoadConfig)
training/verl/third_party/vllm/vllm_v_0_5_4/model_loader.py:175
↓ 4 callersFunction_broadcast_tensor
broadcast tensor across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:131
↓ 4 callersFunction_broadcast_tp_shard_tensor
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:170
↓ 4 callersFunction_check_if_gpu_supports_dtype
(torch_dtype: torch.dtype)
training/verl/third_party/vllm/vllm_v_0_3_1/worker.py:306
↓ 4 callersMethod_forward_micro_batch
(self, module, micro_batch, prompt_length, no_grad=False)
training/verl/workers/actor/dp_prime.py:78
↓ 4 callersFunction_get_cpu_tensor
(tensor: torch.Tensor)
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:124
↓ 4 callersFunction_get_gpt_model
(model)
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:63
↓ 4 callersFunction_is_non_local
(path: str)
training/verl/utils/hdfs_io.py:143
↓ 4 callersFunction_replace_name
(megatron_name, name_mapping)
training/verl/third_party/vllm/vllm_v_0_6_3/megatron_weight_loaders.py:147
↓ 4 callersFunction_replace_name
(megatron_name, name_mapping)
training/verl/third_party/vllm/vllm_v_0_5_4/megatron_weight_loaders.py:146
↓ 4 callersFunction_replace_name
(megatron_name, name_mapping)
training/verl/third_party/vllm/vllm_v_0_4_2/megatron_weight_loaders.py:146
↓ 4 callersFunction_split_args_kwargs_data_proto
(chunks, *args, **kwargs)
training/verl/single_controller/base/decorator.py:45
↓ 4 callersMethodabs_group
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:3739
↓ 4 callersMethodaccent
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4171
↓ 4 callersMethodargs
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5828
↓ 4 callersMethodatom_expr_no_supexpr
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4225
↓ 4 callersMethodbackward
(ctx, grad_output: torch.Tensor)
training/verl/utils/megatron/tensor_parallel.py:118
↓ 4 callersMethodbinom
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4697
↓ 4 callersMethodbuffer_size
(self)
training/verl/utils/dataset/rl_dataset.py:187
↓ 4 callersFunctionbuild_memory_reference_from_module
(module: torch.nn.Module, memory_buffers: Dict[torch.dtype, MemoryBuffe
training/verl/utils/memory_buffer.py:97
↓ 4 callersMethodceil_group
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:4049
↓ 4 callersFunctioncodegen_metrics
( samples_list, generations_list, k_list=[1, 5, 10, 20, 40, 50, 75, 100, 125, 150, 200, 500, 1000]
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/evaluation/compute_code_generation_metrics.py:155
↓ 4 callersMethodcompute_log_prob
Compute logits given a batch of data. Args: data (DataProto): a batch of data represented by DataProto. It must contain key ```in
training/verl/workers/actor/base.py:39
↓ 4 callersFunctionconvert_mp
(mp)
eval/Math/Qwen25-Math/evaluation/latex2sympy/latex2sympy2.py:396
↓ 4 callersMethodeval_at_sub
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2676
↓ 4 callersMethodeval_at_sup
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2741
↓ 4 callersFunctionextract_answer
(pred_str, data_name, use_last_number=True)
eval/Math/Qwen25-Math/evaluation/parser.py:499
↓ 4 callersMethodfloor_group
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:3913
↓ 4 callersMethodfrom_single_dict
(cls, data: Dict[str, Union[torch.Tensor, np.ndarray]], meta_info=None)
training/verl/protocol.py:174
↓ 4 callersMethodfunc_normal_multi_arg
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5180
↓ 4 callersMethodfunc_normal_single_arg
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:5112
↓ 4 callersMethodget
Return a tensor with the input `shape` as a view into the 1-D data starting at `start_index`.
training/verl/utils/megatron/memory.py:33
↓ 4 callersMethodget
(self)
training/verl/utils/rendezvous/ray_backend.py:30
↓ 4 callersFunctionget_args
()
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/runner/parser.py:8
↓ 4 callersMethodget_evaluation_sample
(self)
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/benchmarks/code_execution.py:51
↓ 4 callersFunctionget_micro_data_parallel_group
()
training/verl/workers/hybrid_engine/megatron_vllm.py:418
↓ 4 callersMethodget_placement_groups
(self, strategy="STRICT_PACK", name=None)
training/verl/single_controller/ray/base.py:64
↓ 4 callersFunctionget_tensor_model_parallel_group
Get the tensor model parallel group the caller rank belongs to.
training/verl/third_party/vllm/vllm_v_0_3_1/parallel_state.py:108
↓ 4 callersFunctionget_tensor_model_parallel_group
Get the tensor model parallel group the caller rank belongs to.
training/verl/third_party/vllm/vllm_v_0_6_3/parallel_state.py:291
↓ 4 callersFunctionget_tensor_model_parallel_group
Get the tensor model parallel group the caller rank belongs to.
training/verl/third_party/vllm/vllm_v_0_5_4/parallel_state.py:282
↓ 4 callersFunctionget_tensor_model_parallel_group
Get the tensor model parallel group the caller rank belongs to.
training/verl/third_party/vllm/vllm_v_0_4_2/parallel_state.py:273
↓ 4 callersFunctionget_tensor_model_parallel_world_size
Return world size for the tensor model parallel group.
training/verl/third_party/vllm/vllm_v_0_4_2/parallel_state.py:279
↓ 4 callersFunctionimport_external_libs
(external_libs=None)
training/verl/utils/import_utils.py:41
↓ 4 callersMethodinit
(self)
training/verl/single_controller/base/dp.py:24
↓ 4 callersFunctionis_digit
(s)
eval/utils/grader.py:97
↓ 4 callersFunctionis_digit
(s)
training/verl/utils/reward_score/evaluation_utils/math_util/grader.py:96
↓ 4 callersFunctionis_digit
(s)
data_preprocessing/math_util/grader.py:97
↓ 4 callersFunctionload_megatron_model_weights
(config, model_config, parallel_model,
training/verl/utils/model.py:253
↓ 4 callersFunctionmat_mul_flat
(lh, rh)
eval/Math/Qwen25-Math/evaluation/latex2sympy/latex2sympy2.py:346
↓ 4 callersMethodmatrix
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1340
↓ 4 callersFunctionmy_equals
(ans_p, ans_l)
eval/Math/Qwen25-Math/evaluation/math_utils.py:129
↓ 4 callersMethodnorm_group
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:3635
↓ 4 callersMethodpostfix
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:2442
↓ 4 callersFunctionprint_model_size
(model: nn.Module, name: str = None)
training/verl/utils/model.py:129
↓ 4 callersMethodprofile_run
(self)
training/verl/third_party/vllm/vllm_v_0_3_1/model_runner.py:235
↓ 4 callersMethodread
(self, *args, **kwargs)
eval/Coding/livecodebench/LiveCodeBench-main/lcb_runner/evaluation/utils_execute.py:156
↓ 4 callersFunctionreduce_metrics
(metrics: dict)
training/verl/trainer/ppo/ray_trainer.py:148
↓ 4 callersFunctionrotate_half
Rotates half the hidden dims of the input.
training/verl/models/llama/megatron/layers/parallel_attention.py:116
↓ 4 callersFunctionstream_jsonl
Parses each jsonl line and yields it as a dictionary
eval/utils/data.py:15
↓ 4 callersMethodsubeq
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:6376
↓ 4 callersFunctionsymbolic_equal
(a, b)
eval/Math/Qwen25-Math/evaluation/grader.py:276
↓ 4 callersMethodtransform_swap
(self)
eval/Math/Qwen25-Math/evaluation/latex2sympy/gen/PSParser.py:1105
↓ 3 callersMethod__init__
(self)
training/verl/single_controller/ray/base.py:440
↓ 3 callersMethod__init__
(self, config)
training/verl/workers/fsdp_workers.py:452
↓ 3 callersMethod__init__
(self, dim, max_position_embeddings=2048, base=10000, device=None)
training/verl/models/llama/megatron/layers/parallel_attention.py:37
↓ 3 callersMethod__init__
(self, load_config: LoadConfig)
training/verl/third_party/vllm/vllm_v_0_4_2/model_loader.py:164
↓ 3 callersFunction_broadcast_tp_shard_tensor
broadcast tensor in tp shards across mp_group
training/verl/models/llama/megatron/checkpoint_utils/llama_loader.py:178
↓ 3 callersMethod_build_param_references
(self, pp_rank, maintain_weight=False)
training/verl/workers/hybrid_engine/megatron_vllm.py:88
↓ 3 callersFunction_concat_data_proto_or_future
(output: List)
training/verl/single_controller/base/decorator.py:129
↓ 3 callersFunction_get_and_verify_dtype
( config: PretrainedConfig, dtype: Union[str, torch.dtype], )
training/verl/third_party/vllm/vllm_v_0_3_1/config.py:475
↓ 3 callersFunction_get_and_verify_max_len
Get and verify the model's maximum length.
training/verl/third_party/vllm/vllm_v_0_3_1/config.py:525
↓ 3 callersFunction_get_gpt_model
(model)
training/verl/models/llama/megatron/checkpoint_utils/llama_saver.py:93
↓ 3 callersFunction_str_is_int
(x: str)
training/verl/utils/reward_score/evaluation_utils/math_util/__init__.py:73
↓ 3 callersFunction_str_is_int
(x: str)
data_preprocessing/math_util/__init__.py:73
↓ 3 callersMethod_validate
(self)
training/verl/trainer/ppo/ray_trainer.py:300
← previousnext →101–200 of 2,604, ranked by callers