MCPcopy Create free account

hub / github.com/Gen-Verse/OpenClaw-RL / functions

Functions10,561 in github.com/Gen-Verse/OpenClaw-RL

↓ 1 callersMethod_emit_errors_to_terminal
(ok_count: int, total: int, errors: list[str])
gui-rl/env_pool_server.py:128
↓ 1 callersFunction_empty_like_if_meta
(tensor: torch.Tensor, *, device: torch.device)
Megatron-LM/megatron/post_training/utils.py:52
↓ 1 callersFunction_empty_like_if_meta
(tensor: torch.Tensor, *, device: torch.device)
Megatron-LM/megatron/training/utils.py:737
↓ 1 callersMethod_enable_true_on_policy_optimizations
(self, args)
slime/slime/backends/fsdp_utils/actor.py:197
↓ 1 callersMethod_encode
Tokenize text input.
Megatron-LM/megatron/training/tokenizer/multimodal_tokenizer.py:215
↓ 1 callersMethod_encode
Tokenize text input, w/o chat template
Megatron-LM/megatron/training/tokenizer/sft_tokenizer.py:147
↓ 1 callersMethod_encode_extra_state
(self, state)
Megatron-LM/megatron/core/extensions/kitchen.py:683
↓ 1 callersFunction_encode_observation_for_generation
Encode a single observation turn that may include images/videos in the content list. Trim out the system/tool preamble added by the chat temp
slime/examples/geo3k_vlm_multi_turn/rollout.py:54
↓ 1 callersFunction_encode_single_message
Tokenize a single message via the chat template (no generation prompt). For Qwen3 the result is byte-identical to the corresponding slice of a
swe-rl/generate_with_swe_trajectory.py:535
↓ 1 callersMethod_ensure_formatter
(self, args: Any)
gui-rl/agents/qwen3vl_reward_agent.py:133
↓ 1 callersMethod_ensure_formatter
(self, args: Any)
swe-rl/swe_prm.py:130
↓ 1 callersFunction_ensure_ifbench_dependencies
Install IFBench requirements the first time the module is imported.
slime/slime/rollout/rm_hub/ifbench.py:47
↓ 1 callersFunction_ensure_ifbench_repo
Clone IFBench repo if needed and ensure it is available on sys.path.
slime/slime/rollout/rm_hub/ifbench.py:19
↓ 1 callersFunction_ensure_metadata_overrides
(value: Any)
slime/slime/utils/eval_config.py:86
↓ 1 callersFunction_ensure_openai_base_url
(args)
swe-rl/generate_with_swe_trajectory.py:554
↓ 1 callersFunction_ensure_stop_token_ids
( tokenizer, sampling_params: Dict[str, Any] )
terminal-rl/inference_client.py:102
↓ 1 callersFunction_ensure_terminal_step_metric
(args)
terminal-rl/rollout_log.py:14
↓ 1 callersFunction_eps_clip_hi
(args: Namespace)
openclaw-combine/openclaw_topk_select_loss.py:115
↓ 1 callersFunction_eps_clip_lo
(args: Namespace)
openclaw-combine/openclaw_topk_select_loss.py:110
↓ 1 callersFunction_estimate_pass_at_k
Estimates pass@k of each problem and returns them in an array.
slime/slime/utils/metric_utils.py:43
↓ 1 callersMethod_exchange_loaded_tensors
( self, ten_metas: List[_ShardedTensorMetadata], sharded_state_dict, checkpoint_dir )
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:204
↓ 1 callersFunction_execute_print_only
(args)
slime/tests/ci/gpu_lock_exec.py:66
↓ 1 callersMethod_execute_python
Execute Python code using the sandbox
toolcall-rl/tool_sandbox.py:348
↓ 1 callersMethod_execute_tool
Execute a tool/action in the environment. Args: env: Tau-bench environment instance action: Action to execut
slime/examples/tau-bench/trainable_agents.py:115
↓ 1 callersMethod_execute_tool_calls
( self, turn_idx: int, tool_call_requests: list[Any], )
terminal-rl/agent_runner.py:237
↓ 1 callersMethod_execute_via_api
Execute via Docker Engine API using docker SDK.
swe-rl/mini-swe-agent/src/minisweagent/environments/docker.py:133
↓ 1 callersMethod_execute_via_subprocess
Execute via `docker exec` subprocess.
swe-rl/mini-swe-agent/src/minisweagent/environments/docker.py:88
↓ 1 callersFunction_extract_archives
Extract every .tar / .tar.gz archive found under *root* into its directory.
Megatron-LM/examples/mimo/data/prepare_video_llava_data.py:11
↓ 1 callersFunction_extract_boxed_score
Extract score from \\boxed{...} in evaluator output.
openclaw-rl/oel/eval/personalization_evaluator.py:86
↓ 1 callersFunction_extract_eval_test_files
(instance: dict)
swe-rl/generate_with_swe_trajectory.py:255
↓ 1 callersMethod_extract_experience_from_session
Extract experience items from a completed session and append to global experience.
openclaw-rl/oel/openclaw_oel_api_server.py:1120
↓ 1 callersFunction_extract_ground_truth
Resolve the ground-truth answer from label or metadata.
slime/examples/geo3k_vlm_multi_turn/env_geo3k.py:250
↓ 1 callersFunction_extract_letter_from_response
Best-effort extraction of the selected option letter from the model response.
slime/slime/rollout/rm_hub/gpqa.py:22
↓ 1 callersFunction_extract_logprobs
(choice: dict[str, Any])
openclaw-tinker/api_server.py:77
↓ 1 callersFunction_extract_logprobs_from_chat_response
(choice: dict[str, Any])
openclaw-rl/openclaw_api_server.py:84
↓ 1 callersFunction_extract_logprobs_from_chat_response
(choice: dict[str, Any])
openclaw-rl/oel/openclaw_oel_api_server.py:168
↓ 1 callersFunction_extract_logprobs_from_chat_response
(choice: dict[str, Any])
openclaw-opd/openclaw_opd_api_server.py:79
↓ 1 callersFunction_extract_logprobs_from_chat_response
(choice: dict[str, Any])
openclaw-fireworks/fireworks_server.py:108
↓ 1 callersFunction_extract_patch_from_submission
(output: str)
swe-rl/generate_with_swe_trajectory.py:187
↓ 1 callersFunction_extract_prm_sign_from_text
(text: str)
terminal-rl/agent/prm_agent.py:45
↓ 1 callersFunction_extract_prm_sign_from_text
(text: str)
gui-rl/agents/qwen3vl_reward_agent.py:68
↓ 1 callersFunction_extract_prm_sign_from_text
(text: str)
toolcall-rl/generate_with_retool.py:333
↓ 1 callersFunction_extract_prm_sign_from_text
(text: str)
swe-rl/swe_prm.py:83
↓ 1 callersFunction_extract_prm_text
(output: Any)
toolcall-rl/generate_with_retool.py:355
↓ 1 callersMethod_extract_session_experience
Extract experience from current session conversation and store per-session. Unlike _extract_experience_from_session which accumulates globall
openclaw-rl/oel/openclaw_oel_api_server.py:1056
↓ 1 callersMethod_extract_tool_call
Parse the latest tool call payload from the assistant response. Supports the <tool_call>{...}</tool_call> convention used in the
slime/examples/geo3k_vlm_multi_turn/env_geo3k.py:57
↓ 1 callersFunction_extract_tool_calls
Parse tool-call tags from assistant text into OpenAI-style tool_calls.
openclaw-tinker/api_server.py:96
↓ 1 callersFunction_fetch
(idx: int, url: str)
terminal-rl/router_server.py:211
↓ 1 callersFunction_fetch_requirements
(path)
slime/setup.py:8
↓ 1 callersFunction_file_lock
(path: str)
gui-rl/desktop_env/providers/volcengine/manager.py:67
↓ 1 callersFunction_filter_headers
Filter out hop-by-hop headers that should not be forwarded.
slime/slime/router/middleware_hub/radix_tree_middleware.py:30
↓ 1 callersMethod_finalize_last_turn
For the last turn in a session, fire teacher task even without next_state.
openclaw-rl/oel/openclaw_oel_api_server.py:887
↓ 1 callersFunction_finalize_sample
(sample: Sample, tokenizer, response_tokens, multimodal_train_inputs_buffer)
slime/examples/geo3k_vlm_multi_turn/rollout.py:300
↓ 1 callersMethod_find_classes
Finds the class folders in a dataset. Args: dir (string): Root directory path. Returns: tuple: (class
Megatron-LM/megatron/legacy/data/image_folder.py:191
↓ 1 callersFunction_find_last_answer_boxed_span
Return (start, end) of the *last* 'Answer: \\boxed{...}' in *text*.
toolcall-rl/generate_with_retool.py:213
↓ 1 callersMethod_find_node
(self, url: str)
swe-rl/server/swe_env_pool_server.py:100
↓ 1 callersMethod_find_node_by_text
Find node by exact text match. Args: text: Text to find Returns: Node if found, None otherwise
slime/slime/router/middleware_hub/radix_tree.py:374
↓ 1 callersMethod_find_outdated_nodes
Find nodes that should be removed based on weight version threshold. Uses layer-by-layer traversal - if parent is outdated, children
slime/slime/router/middleware_hub/radix_tree.py:473
↓ 1 callersMethod_find_pattern_indices
( self, template, pattern, start_idx=0, allow_first_mismatch=False )
Megatron-LM/examples/mimo/data/energon_avlm_task_encoder.py:156
↓ 1 callersMethod_find_pattern_indices
( self, template, pattern, start_idx=0, allow_first_mismatch=False )
Megatron-LM/examples/mimo/data/energon_vlm_task_encoder.py:135
↓ 1 callersMethod_finish_capturing
Finish capturing CUDA Graphs and clean up the related state.
Megatron-LM/megatron/core/transformer/cuda_graphs.py:1872
↓ 1 callersMethod_finish_url
Mark the request to the given URL as finished
slime/slime/router/router.py:242
↓ 1 callersMethod_fire_evaluation_task
(self, session_id, turn_num, turn_data, next_state)
openclaw-tinker/api_server.py:870
↓ 1 callersMethod_fire_opd_task
(self, session_id: str, turn_num: int, turn_data: dict[str, Any], next_state: dict[str, Any])
openclaw-opd/openclaw_opd_api_server.py:693
↓ 1 callersMethod_fire_opd_task
( self, session_id: str, turn_num: int, turn_data: dict[str, Any], next_state: dict[str, Any],
openclaw-fireworks/fireworks_server.py:818
↓ 1 callersMethod_fire_opd_task
(self, session_id, turn_num, turn_data, next_state)
openclaw-tinker/api_server.py:667
↓ 1 callersMethod_fire_prm_scoring
(self, session_id: str, turn_num: int, response_text: str, next_state)
openclaw-rl/openclaw_api_server.py:457
↓ 1 callersMethod_fire_prm_scoring
(self, session_id, turn_num, response_text, next_state)
openclaw-tinker/api_server.py:512
↓ 1 callersMethod_fire_teacher_task
Fire async teacher evaluation for a turn.
openclaw-rl/oel/openclaw_oel_api_server.py:872
↓ 1 callersFunction_first_not_missing
(*values: Any)
slime/slime/utils/eval_config.py:62
↓ 1 callersFunction_fix_a_slash_b
(string)
slime/slime/rollout/rm_hub/math_utils.py:61
↓ 1 callersFunction_fix_fracs
(string)
slime/slime/rollout/rm_hub/math_utils.py:30
↓ 1 callersFunction_fix_sqrt
(string)
slime/slime/rollout/rm_hub/math_utils.py:84
↓ 1 callersMethod_fix_tensor_parallel_attributes
(self, module)
Megatron-LM/megatron/core/distributed/fsdp/mcore_fsdp_adapter.py:155
↓ 1 callersMethod_forward
(self, *inputs)
Megatron-LM/megatron/core/pipeline_parallel/utils.py:206
↓ 1 callersMethod_forward_impl
(self, input, weight, *args, **kwargs)
Megatron-LM/megatron/core/tensor_parallel/layers.py:964
↓ 1 callersMethod_forward_impl
(self, input, weight, *args, **kwargs)
Megatron-LM/megatron/core/tensor_parallel/layers.py:1256
↓ 1 callersFunction_fp8_create_transpose_cache_fallback
(tensors: List[torch.Tensor])
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:196
↓ 1 callersFunction_fp8_quantize_fallback
( model_params: List[torch.Tensor], main_params: List[torch.Tensor], start_offsets: List[int],
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:269
↓ 1 callersFunction_free_up_some_gpu_memory
()
Megatron-LM/tools/checkpoint/checkpoint_inspector.py:383
↓ 1 callersFunction_fro_norm
Frobenius norm; supports sharded tensors (sum of shard ||·||_F^2).
Megatron-LM/tests/functional_tests/python_test_utils/test_optimizer_grads_match.py:17
↓ 1 callersFunction_fsdp_compute_log_probs
Compute log P(token | context) without Megatron fused kernel.
openclaw-rl/oel/oel_distillation_loss.py:69
↓ 1 callersFunction_fsdp_get_responses
Yield (logits_chunk[R,V], tokens_chunk[R]) per sample -- FSDP version.
openclaw-rl/oel/oel_distillation_loss.py:49
↓ 1 callersMethod_gather_at_indices_multi_cand
K-loop teacher gather-at-indices for ``--distill-subset-mode student``. ``teacher_tokens_cand[i]`` holds K_i hint-enhanced token sequences
slime/slime/backends/megatron_utils/actor.py:1012
↓ 1 callersFunction_gather_student_raw_logits_at_indices
Gather raw global logits at ``indices`` (autograd-connected).
openclaw-combine/hint_opd_loss.py:146
↓ 1 callersMethod_gather_var
Compute statistic across the non-dummy heads.
Megatron-LM/examples/multimodal/nvlm/internvit.py:113
↓ 1 callersFunction_generate_impl_trajectory
(args, sample: Sample, sampling_params: dict)
swe-rl/generate_with_swe_trajectory.py:901
↓ 1 callersFunction_generate_mac_address
()
gui-rl/desktop_env/providers/vmware/manager.py:67
↓ 1 callersMethod_get_all_ranks_time_string
Report times across all ranks.
Megatron-LM/megatron/core/timers.py:346
↓ 1 callersMethod_get_api_client
(self)
swe-rl/mini-swe-agent/src/minisweagent/environments/docker.py:112
↓ 1 callersMethod_get_async_caller
(self)
Megatron-LM/megatron/core/dist_checkpointing/strategies/async_utils.py:521
↓ 1 callersMethod_get_available_models_list
Returns a list of available megatron tokenizers.
Megatron-LM/megatron/core/tokenizers/text/libraries/megatron_hf_tokenizer.py:144
↓ 1 callersFunction_get_block_submodules
Retrieve or construct TransformerBlockSubmodules based on the provided specification. Args: config (TransformerConfig): Configuratio
Megatron-LM/megatron/core/transformer/transformer_block.py:228
↓ 1 callersFunction_get_candidate_prefixes
(keys: list[str])
slime/tools/convert_fsdp_to_hf.py:73
↓ 1 callersFunction_get_checkpoint_format
Get the format of an existing checkpoint.
Megatron-LM/megatron/training/checkpointing.py:1069
↓ 1 callersFunction_get_class_from_path
(subdomains, imp)
Megatron-LM/tests/unit_tests/test_imports.py:43
↓ 1 callersFunction_get_cleared_task_keys
()
gui-rl/generate_with_gui.py:56
↓ 1 callersFunction_get_config
Get various params from dense state dict and moe model.
Megatron-LM/megatron/core/transformer/moe/upcycling_utils.py:32
↓ 1 callersFunction_get_content_text
(entry: dict)
swe-rl/mini-swe-agent/src/minisweagent/models/utils/cache_control.py:1
↓ 1 callersFunction_get_cuda_bare_metal_version
(cuda_dir)
Megatron-LM/megatron/legacy/fused_kernels/__init__.py:57
← previousnext →2,801–2,900 of 10,561, ranked by callers