MCPcopy Create free account

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

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

Methodread
Read bytes into a numpy array. Let `size` be the `count` * `DType.size(dtype)`. If the requested span of bytes [`offset`, `offset` +
Megatron-LM/megatron/core/datasets/indexed_dataset.py:484
Methodread
(self, dtype: Type[numpy.number], count: int, offset: int)
Megatron-LM/megatron/core/datasets/indexed_dataset.py:549
Methodread_data_into_queue
()
slime/slime_plugins/rollout_buffer/generator/base_generator.py:195
Methodread_frame
(self, video_path, bound=None, fps=2)
Megatron-LM/examples/multimodal/evaluation/evaluation_datasets.py:1276
Functionread_golden_values_from_json
( golden_values_path: Union[str, pathlib.Path] )
Megatron-LM/tests/functional_tests/python_test_utils/common.py:161
Methodread_metadata
(self)
slime/tools/convert_fsdp_to_hf.py:27
Methodread_metadata
(self)
slime/tools/convert_torch_dist_to_hf.py:36
Functionread_tb_logs_as_list
Reads a TensorBoard Events file from the input path, and returns the summary specified as input as a list. Args: path: str, path to t
Megatron-LM/tests/functional_tests/python_test_utils/common.py:94
Methodread_video_ours
(self, video_path, bound=None)
Megatron-LM/examples/multimodal/evaluation/evaluation_datasets.py:1285
Functionrebuild_tokenizer
(args)
Megatron-LM/megatron/training/global_vars.py:164
Methodreceive_lm
Receive LM model parameters over queue and save them in self.models
Megatron-LM/tools/checkpoint/saver_base.py:353
Methodreceive_model
(self)
Megatron-LM/tools/checkpoint/saver_core.py:53
Methodreceive_model
(self)
Megatron-LM/tools/checkpoint/saver_llava.py:359
Methodrecord_model_turn
( self, turn_idx: int, *, assistant_text: str, tool_calls: Optional[Li
terminal-rl/agent/prm_agent.py:109
Methodrecord_policy_turn
(self, *, action_text: str, response: str, screenshot_bytes: bytes)
gui-rl/agents/qwen3vl_agent.py:315
Methodrecord_tool_result
(self, tool_call_request: Any, raw_result: Any)
terminal-rl/rollout_agent.py:25
Methodrecord_tool_result
( self, turn_idx: int, tool_call_request, raw_result: Any )
terminal-rl/agent/prm_agent.py:125
Methodrecord_tool_result
(self, tool_call_request: Any, raw_result: Any)
terminal-rl/agent/camel_agent.py:550
Methodrecover_rollout_engines
Restart any dead rollout engines and update num_new_engines for update_weights detection.
slime/slime/ray/rollout.py:200
Methodrecv_anysource
(self, *a, **kw)
slime/slime/utils/reloadable_process_group.py:238
Functionreduce_scatter_last_dim_to_tensor_parallel_region
Wrapper for autograd function: forward: RS, backward AG: AG <last dim>
Megatron-LM/megatron/core/tensor_parallel/mappings.py:529
Methodreduce_scatter_tensor_coalesced
(self, *a, **kw)
slime/slime/utils/reloadable_process_group.py:223
Methodregister_subclass
Register subclass for unwrapping.
Megatron-LM/megatron/rl/__init__.py:47
Methodregister_submesh
Register a submesh with identifier: (*submesh, is_expert_parallel) in the mesh library.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:499
Methodrelease_memory_occupation
(self)
slime/slime/backends/sglang_utils/sglang_engine.py:358
Methodrelease_modifier_keys
()
gui-rl/agents/qwen35_agent.py:549
Methodrelease_module_parameters
(module, bwd, *unused)
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/megatron_fsdp.py:501
Methodrelease_params_fp8_transpose_cache
(params)
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/megatron_fsdp.py:508
Methodrelease_state
Release reference, this helps avoid memory leak.
Megatron-LM/megatron/core/models/common/model_chunk_schedule_plan.py:424
Methodreload_model_params
Refreshes any internal state from the current model parameters. Call whenever the parameters are changed outside of the optimizer. For
Megatron-LM/megatron/core/optimizer/optimizer.py:251
Methodreload_model_params
(self, state_dict=None)
Megatron-LM/megatron/core/optimizer/optimizer.py:492
Methodreload_model_params
(self, state_dict=None)
Megatron-LM/megatron/core/optimizer/optimizer.py:1173
Methodremaining_prompt_length
Get the length of the remaining prompt tokens.
Megatron-LM/megatron/core/inference/inference_request.py:254
Methodremove_extra_states_check
Remove _extra_state from unexpected keys. These keys are for dist ckpt compatibility with SequentialMLP.
Megatron-LM/megatron/core/transformer/moe/experts.py:222
Methodremove_sharded_tensors
Removes checkpoint files whose keys have the given prefix. Performs the following steps: 1. checks whether there are files that start
Megatron-LM/megatron/core/dist_checkpointing/strategies/torch.py:1035
Functionreplace_press_less_than
(match)
gui-rl/desktop_env/desktop_env.py:45
Methodreplace_screen_env_in_command
(command)
gui-rl/desktop_env/controllers/setup.py:355
Methodreq_count
Returns the total number of requests.
Megatron-LM/megatron/core/inference/batch_dimensions_utils.py:120
Methodrequest_id
Get request id. Returns: (int) Request id.
Megatron-LM/megatron/core/inference/inference_request.py:404
Functionrequest_thread
()
swe-rl/mini-swe-agent/tests/agents/test_interactive_textual.py:578
Functionreset
(request: Request)
terminal-rl/router_server.py:345
Functionreset
(request: Request)
terminal-rl/remote/pool_server.py:427
Functionreset
()
gui-rl/env_pool_server.py:368
Methodreset
( self, *, task_meta: dict[str, Any], run_ctx: RunContext, timeouts: T
terminal-rl/remote/terminal_env.py:85
Methodreset
(self, task_config: Optional[Dict[str, Any]] = None, seed=None, options=None)
gui-rl/desktop_env/desktop_env.py:242
Methodreset
(self, _logger=None)
gui-rl/agents/qwen35_agent.py:101
Methodreset
(self, _logger=None)
gui-rl/agents/qwen3vl_agent.py:85
Methodreset
Reset timer.
Megatron-LM/megatron/core/timers.py:60
Methodreset
(self)
Megatron-LM/megatron/core/timers.py:90
Methodreset
Handle the processing tasks and reset the pipeline.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:2930
Methodreset
Reset the pipeline state.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:3319
Methodreset
Reset the allocator to initial state. This resets the available block count to the entire memory pool (except for the dummy block).
Megatron-LM/megatron/core/inference/contexts/dynamic_block_allocator.py:112
Methodreset
Resets the inference state for a new batch.
Megatron-LM/megatron/core/inference/contexts/static_context.py:74
Methodreset
Resets all Mamba states and frees all allocated slots.
Megatron-LM/megatron/core/inference/contexts/attention_context/mamba_metadata.py:54
Methodreset
(self)
Megatron-LM/megatron/core/inference/contexts/attention_context/mha_metadata.py:170
Methodreset
Reset the metadata.
Megatron-LM/megatron/core/inference/contexts/attention_context/metadata_base.py:26
Methodreset
(self)
slime/examples/geo3k_vlm_multi_turn/base_env.py:6
Methodreset_drop_path
(self, drop_path_rate)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:275
Functionreset_env_vars
Reset environment variables
Megatron-LM/tests/unit_tests/conftest.py:100
Methodreset_eval_scores
(self)
openclaw-fireworks/fireworks_server.py:871
Methodreset_eval_scores
(self)
openclaw-tinker/api_server.py:345
Functionreset_global_stats
Reset global model stats and ensure exclusive access for tests that need it. This fixture should be used by any test that depends on global model
swe-rl/mini-swe-agent/tests/conftest.py:14
Methodreset_index
Delete existing index and create a new
Megatron-LM/megatron/legacy/data/realm_index.py:166
Functionresolve_artifact_config
(cluster: str)
Megatron-LM/tests/test_utils/python_scripts/recipe_parser.py:37
Functionresolve_cluster_config
(cluster: str)
Megatron-LM/tests/test_utils/python_scripts/recipe_parser.py:23
Methodresolve_tensor
(self, read_item)
Megatron-LM/tools/checkpoint/checkpoint_inspector.py:326
Methodresolve_tensor
Override to add FP8 support. Narrowing the Float8Tensor can create incontiguous tensors and there are no `copy` kernels for such case
Megatron-LM/megatron/core/dist_checkpointing/strategies/torch.py:656
Methodrestore
(self, tag: str)
slime/slime/utils/tensor_backper.py:69
Methodrestore
(self, tag: str)
slime/slime/utils/tensor_backper.py:99
Methodrestore_custom_attrs
(module, custom_attrs)
Megatron-LM/megatron/core/distributed/torch_fully_sharded_data_parallel.py:102
Methodrestore_from_cpu
Function used for RL training. Restore optimizer state tensors from CPU back to GPU for training.
Megatron-LM/megatron/core/optimizer/optimizer.py:364
Methodrestore_tensor_device
Restores all tensors to their original devices, if a move is required. Using non_blocking=True allows for asynchronous copying.
Megatron-LM/megatron/core/dist_checkpointing/tensor_aware_state_dict.py:264
Methodresume
Resume/start energy monitor.
Megatron-LM/megatron/core/energy_monitor.py:54
Methodresume
(self)
Megatron-LM/megatron/rl/inference/megatron.py:259
Methodresume
(self)
Megatron-LM/megatron/rl/server/inference/inference_interface_server.py:96
Methodresume_memory_occupation
Available tags for multi-stage resume: weights, kv_cache
slime/slime/backends/sglang_utils/sglang_engine.py:362
Functionretrieve_endpoint
Endpoint that accepts queries and performs retrieval. Input format: { "queries": ["What is Python?", "Tell me about neural network
slime/examples/search-r1/local_dense_retriever/retrieval_server.py:329
Methodretrieve_from_text
Get token information from text input
slime/slime/router/router.py:203
Functionreturn_none
(input_tensor)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:487
Methodreturns
Return a single information on function return. Takes the first return information.
Megatron-LM/tools/retro/config_utils.py:190
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/azure/provider.py:110
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/aws/provider.py:103
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/docker/provider.py:153
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/virtualbox/provider.py:109
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/volcengine/provider.py:103
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/vmware/provider.py:93
Methodrevert_to_snapshot
(self, path_to_vm: str, snapshot_name: str)
gui-rl/desktop_env/providers/aliyun/provider.py:168
Functionreward_func
(args, sample_or_samples, **kwargs)
openclaw-rl/openclaw_api_server.py:160
Functionreward_func
(args, sample_or_samples, **kwargs)
openclaw-rl/oel/openclaw_oel_api_server.py:234
Functionreward_func
(args, sample_or_samples, **kwargs)
openclaw-opd/openclaw_opd_api_server.py:225
Functionreward_func
(args, sample: Sample | list[Sample], **kwargs)
gui-rl/generate_with_gui.py:734
Functionreward_func
Reward is pre-set in generate(); this is a no-op pass-through.
gui-rl/generate_with_agentnet.py:280
Functionreward_func
Tool call reward function using math_dapo as primary reward model
toolcall-rl/generate_with_retool.py:856
Functionreward_func
(args, sample, **kwargs)
swe-rl/generate_with_swe.py:29
Functionreward_func
(args, sample, **kwargs)
slime/examples/on_policy_distillation/on_policy_distillation.py:8
Functionreward_func
Reward function using math_dapo scoring.
slime/examples/strands_sglang/generate_with_strands.py:100
Functionrl_inference_interface_shutdown
()
Megatron-LM/megatron/rl/rl_utils.py:2342
Functionrng_context_wrapper
Wrapper to add rng context to submodule callables
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:564
Methodrollout
(self, request: RolloutRequest)
Megatron-LM/megatron/rl/server/agent/fastapi_env_server.py:141
Methodrollout
(self, request: RolloutRequest)
Megatron-LM/megatron/rl/agent/weighted_multi_task.py:154
← previousnext →8,501–8,600 of 10,561, ranked by callers