MCPcopy Create free account

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

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

Methodstart
Start the health monitor thread. Called once during initialization. Returns: True if the monitor was started, False if there are
slime/slime/utils/health_monitor.py:36
Methodstart_emulator
(self, path_to_vm: str, headless: bool, os_type: str = None, *args, **kwargs)
gui-rl/desktop_env/providers/azure/provider.py:35
Methodstart_emulator
(self, path_to_vm: str, headless: bool, *args, **kwargs)
gui-rl/desktop_env/providers/aws/provider.py:24
Methodstart_emulator
(self, path_to_vm: str, headless: bool, os_type: str)
gui-rl/desktop_env/providers/docker/provider.py:87
Methodstart_emulator
(self, path_to_vm: str, headless: bool, os_type: str = None, *args, **kwargs)
gui-rl/desktop_env/providers/virtualbox/provider.py:58
Methodstart_emulator
(self, path_to_vm: str, headless: bool, *args, **kwargs)
gui-rl/desktop_env/providers/volcengine/provider.py:35
Methodstart_emulator
(self, path_to_vm: str, headless: bool, os_type: str)
gui-rl/desktop_env/providers/vmware/provider.py:47
Methodstart_emulator
(self, path_to_vm: str, headless: bool, *args, **kwargs)
gui-rl/desktop_env/providers/aliyun/provider.py:41
Methodstart_grad_sync
Initiates grad sync (all-reduce or reduce-scatter) communication operations for all model gradients. When overlap_grad_reduc
Megatron-LM/megatron/core/distributed/data_parallel_base.py:34
Methodstart_grad_sync
Initiates grad sync (all-reduce or reduce-scatter) communication operations for all model gradients. When overlap_grad_reduc
Megatron-LM/megatron/core/distributed/distributed_data_parallel.py:510
Methodstart_layer
(self)
slime/slime/backends/sglang_utils/qwen3_5.py:118
Methodstart_method
This method adds the start timers. Both cuda event and perf_counter are added. If bdata is set to true from __call__, this method ski
Megatron-LM/megatron/core/utils.py:1471
Methodstart_param_sync
Initiates param sync (all-gather) communication operations for all model parameters. By default, when overlap_param_gather is set to
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/megatron_fsdp.py:1015
Methodstart_profile
( self, # The output directory output_dir: str | None = None, # If set, it pro
slime/slime/backends/sglang_utils/sglang_engine.py:445
Functionstart_recording
()
gui-rl/env_pool_server.py:431
Functionstart_recording
()
gui-rl/desktop_env/server/main.py:1495
Methodstart_recording
Starts recording the screen.
gui-rl/desktop_env/controllers/python.py:410
Functionstart_rollout
(request: Request, background: BackgroundTasks)
slime/slime_plugins/rollout_buffer/buffer.py:326
Methodstart_turn_loop
( self, input_message: Union[BaseMessage, str], response_format: Optional[Type[BaseMod
terminal-rl/agent/camel_agent.py:363
Methodstate
Gets the model chunk state.
Megatron-LM/megatron/core/models/common/model_chunk_schedule_plan.py:420
Methodstate
Return optimizer state with tuple keys, where the first element is the index of the optimizer in the list of chained optimizers.
Megatron-LM/megatron/core/optimizer/optimizer.py:1126
Methodstate_dict
Method that returns a state dict to be checkpointed. Args: data_iterator: the data iterator that needs to be checkpointed (or Non
Megatron-LM/megatron/core/rerun_state_machine.py:719
Methodstate_dict
Method to capture the state of the iterator as a serializable dict.
Megatron-LM/megatron/core/rerun_state_machine.py:1079
Methodstate_dict
Method to capture the state of the error injector as a serializable dict.
Megatron-LM/megatron/core/rerun_state_machine.py:1260
Methodstate_dict
Return state_dict.
Megatron-LM/megatron/core/optimizer/optimizer.py:268
Methodstate_dict
(self)
Megatron-LM/megatron/core/optimizer/optimizer.py:1178
Methodstate_dict
(self)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:30
Methodstate_dict
(self)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:46
Methodstate_dict
(self)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:132
Methodstate_dict
(self)
slime/slime/backends/fsdp_utils/checkpoint.py:39
Methodstate_dict
(self)
slime/slime/backends/fsdp_utils/checkpoint.py:55
Methodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
Megatron-LM/megatron/legacy/model/classification.py:76
Methodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
Megatron-LM/megatron/legacy/model/realm_model.py:185
Methodstate_dict_for_save_checkpoint
(self, prefix='', keep_vars=False)
Megatron-LM/megatron/legacy/model/gpt_model.py:104
Methodstate_dict_for_save_checkpoint
For easy load.
Megatron-LM/megatron/legacy/model/language_model.py:571
Methodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
Megatron-LM/megatron/legacy/model/biencoder_model.py:304
Methodstate_dict_for_save_checkpoint
Use this function to override the state dict for saving checkpoints.
Megatron-LM/megatron/legacy/model/module.py:31
Methodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
Megatron-LM/megatron/legacy/model/bert_model.py:222
Methodstate_dict_for_save_checkpoint
For easy load when model is combined with other heads, add an extra key.
Megatron-LM/megatron/legacy/model/multiple_choice.py:87
Methodstate_dict_for_save_checkpoint
Returns wrapped module's state_dict for checkpoint saving.
Megatron-LM/megatron/core/distributed/data_parallel_base.py:84
Methodstate_dict_for_save_checkpoint
Override state dict for saving checkpoints Use this function to override the state dict for saving checkpoints. Args: pre
Megatron-LM/megatron/core/transformer/module.py:43
Methodstate_dict_for_save_checkpoint
Retrieve state_dict from the module being wrapped.
Megatron-LM/megatron/core/transformer/module.py:470
Functionstate_restore_func
(state_dict)
Megatron-LM/megatron/training/initialize.py:101
Functionstate_save_func
()
Megatron-LM/megatron/training/initialize.py:98
Functionstatic_inference_engine
(gpt2_tiktoken_tokenizer)
Megatron-LM/tests/unit_tests/test_inference.py:26
Functionstatus
()
terminal-rl/router_server.py:205
Functionstatus
()
terminal-rl/remote/pool_server.py:370
Functionstatus
()
gui-rl/env_pool_server.py:324
Functionstatus
()
swe-rl/server/swe_exec_server.py:105
Functionstatus
()
swe-rl/server/swe_env_pool_server.py:226
Functionstep
()
gui-rl/env_pool_server.py:399
Methodstep
(self, action, pause=2)
gui-rl/desktop_env/desktop_env.py:381
Methodstep
(self)
swe-rl/mini-swe-agent/src/minisweagent/agents/interactive.py:77
Methodstep
Override step to provide progress updates.
swe-rl/mini-swe-agent/src/minisweagent/run/extra/swebench.py:61
Methodstep
Step the optimizer.
Megatron-LM/megatron/core/optimizer/optimizer.py:302
Methodstep
(self)
Megatron-LM/megatron/core/optimizer/optimizer.py:593
Methodstep
ChainedOptimizer will step all optimizers one by one.
Megatron-LM/megatron/core/optimizer/optimizer.py:1308
Methodstep
(self, response_text: str)
slime/examples/geo3k_vlm_multi_turn/base_env.py:9
Methodstep_legacy
Synchronous wrapper for `self.async_step`.
Megatron-LM/megatron/core/inference/engines/dynamic_engine.py:1242
Methodstep_with_ready_grads
Step the optimizer with ready gradients, return successful.
Megatron-LM/megatron/core/optimizer/optimizer.py:190
Methodstep_with_ready_grads
Step the optimizer with ready gradients, return successful. Under the hood, either launch synchronous param all-gathers or get ready to launch
Megatron-LM/megatron/core/optimizer/distrib_optimizer.py:2582
Methodstop
(self)
openclaw-rl/openclaw_api_server.py:745
Methodstop
(self)
openclaw-rl/oel/openclaw_oel_api_server.py:1677
Methodstop
(self)
openclaw-opd/openclaw_opd_api_server.py:1016
Methodstop
(self)
openclaw-tinker/api_server.py:459
Methodstop
Stop the timer. Args: barrier (bool, optional): Synchronizes ranks before stopping. Defaults to False.
Megatron-LM/megatron/core/timers.py:51
Methodstop
(self, barrier=False)
Megatron-LM/megatron/core/timers.py:87
Methodstop
(self)
slime/slime/utils/profile_utils.py:124
Methodstop
(self)
slime/slime/utils/profile_utils.py:145
Methodstop
Stop the health monitor thread completely. Called during dispose.
slime/slime/utils/health_monitor.py:62
Methodstop_emulator
(self, path_to_vm, region=None)
gui-rl/desktop_env/providers/azure/provider.py:184
Methodstop_emulator
(self, path_to_vm, region=None)
gui-rl/desktop_env/providers/aws/provider.py:259
Methodstop_emulator
(self, path_to_vm: str, region=None, *args, **kwargs)
gui-rl/desktop_env/providers/virtualbox/provider.py:118
Methodstop_emulator
(self, path_to_vm, region=None)
gui-rl/desktop_env/providers/volcengine/provider.py:136
Methodstop_emulator
(self, path_to_vm: str, region=None, *args, **kwargs)
gui-rl/desktop_env/providers/vmware/provider.py:100
Methodstop_emulator
(self, path_to_vm: str, region: str = None)
gui-rl/desktop_env/providers/aliyun/provider.py:201
Functionstop_global_worker
()
openclaw-rl/openclaw_rollout.py:26
Functionstop_global_worker
()
openclaw-rl/oel/openclaw_oel_rollout.py:35
Functionstop_global_worker
()
openclaw-opd/openclaw_opd_rollout.py:26
Functionstop_global_worker
()
openclaw-combine/openclaw_combine_rollout.py:27
Functionstop_global_worker
()
openclaw-combine/openclaw_combine_select_rollout.py:41
Methodstop_method
This method adds the stop timers. Both cuda event and perf_counter are added. If bdata is set to true from __call__, this method skip
Megatron-LM/megatron/core/utils.py:1505
Methodstop_profile
(self)
slime/slime/backends/sglang_utils/sglang_engine.py:474
Functionstr_to_reduction_enum
str -> EntropyReductionEnum
Megatron-LM/megatron/core/fusions/linear_cross_entropy/utils.py:17
Methodstream_token
Asynchronously streams a token for the given request.
Megatron-LM/megatron/core/inference/text_generation_controllers/text_generation_controller.py:1514
Methodstream_tokens
Asynchronously streams tokens for the given requests. Args: sampling_params (SamplingParams): The sampling parameters.
Megatron-LM/megatron/core/inference/text_generation_controllers/text_generation_controller.py:1486
Methodsub_optimizers
Return the list of sub-optimizers.
Megatron-LM/megatron/core/optimizer/cpu_offloading/hybrid_optimizer.py:466
Methodsubmit_step_judge
( self, args: Any, *, problem_statement: str, step_debug: list[dict],
swe-rl/swe_prm.py:388
Functionsubmodule_attn_forward
Performs same attnention forward logic as GPT Model.
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:354
Functionsubmodule_combine_forward
# Triggers token combine and the remaining computation in the transformer layer. # The `mlp_bda` computation is placed after `mlp.com
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:444
Functionsubmodule_dispatch_forward
Dispatches tokens to the experts based on the router output.
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:397
Functionsubmodule_moe_forward
Run forward pass for computations between dispatch and combine: post dispatch->experts->combine preprocess
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:413
Functionsubmodule_mtp_attn_forward
(node, hidden_states)
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:521
Functionsubmodule_mtp_postprocess_forward
(node, hidden_states)
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:556
Functionsubmodule_post_attn_forward
Run forward pass for computations between attention and dispatch: pre mlp layernorm->router->dispatch preprocess
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:369
Methodsucceeded
Request experienced no non-transient errors.
Megatron-LM/megatron/core/inference/inference_request.py:361
Functionsum_of_token
(x: torch.Tensor)
slime/slime/backends/megatron_utils/cp_utils.py:75
Methodsummarize_load_times
Summarize load times.
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:119
Methodsuspend
Suspend request by storing references to previous prompt, generations, and sampling params. Args: tokenizer (MegatronToke
Megatron-LM/megatron/core/inference/inference_request.py:412
Methodsuspend
(self)
Megatron-LM/megatron/rl/inference/megatron.py:254
← previousnext →9,001–9,100 of 10,561, ranked by callers