MCPcopy Create free account

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

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

Functionforward_step
Forward step.
Megatron-LM/pretrain_vision_inpaint.py:80
Functionforward_step
Forward step.
Megatron-LM/pretrain_vision_classify.py:64
Functionforward_step
Forward step.
Megatron-LM/tools/retro/sft/sft_retro.py:164
Functionforward_step
Forward step for MIMO model training. Args: data_iterator: iterator over the dataset model: MIMO model instance Returns:
Megatron-LM/examples/mimo/train.py:161
Functionforward_step
Forward training step. Args: data_iterator: Input data iterator model: The GPT Model
Megatron-LM/examples/post_training/modelopt/finetune.py:465
Functionforward_step
Forward step.
Megatron-LM/examples/academic_paper_scripts/detxoify_lm/finetune_gpt.py:83
Functionforward_step
Forward training step. Args: data_iterator (torch.utils.data.dataloader): Input data iterator model: Multimodal model Return
Megatron-LM/examples/multimodal/train.py:254
Functionforward_step
Forward step used by Megatron's pipeline engine. Args: data_iterator (DataIterator): Input data iterator. model (GPTM
slime/slime/backends/megatron_utils/model.py:190
Methodforward_step
Forward step. Args: orig_input_ids (T.Iterator): Input token IDs. model (GPTModel): The GPT Model.
Megatron-LM/tests/functional_tests/test_cases/common/ckpt_converter/__main__.py:321
Methodforward_step
Forward step. Args: orig_input_ids (T.Iterator): Input token IDs. model (GPTModel): The GPT Model.
Megatron-LM/tests/functional_tests/test_cases/common/ckpt_converter/__main__.py:686
Functionforward_step_func
Forward training step. Copied from `pretrain_gpt.py`
Megatron-LM/tests/unit_tests/pipeline_parallel/test_pipeline_layout.py:234
Functionforward_step_func
(data_iterator, model)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:138
Functionforward_step_func
(data_iterator, model)
Megatron-LM/tests/unit_tests/transformer/test_full_cuda_graph.py:31
Functionforward_tp_epilogue
forward epilogue in tp
Megatron-LM/megatron/core/fusions/linear_cross_entropy/blackwell/triton.py:136
Functionforward_tp_epilogue_update_logprobs
update logprobs in tp
Megatron-LM/megatron/core/fusions/linear_cross_entropy/blackwell/triton.py:205
Methodfp32_param_copy_back_gpu_hook
(optimizer, args, kwargs)
Megatron-LM/megatron/core/optimizer/cpu_offloading/hybrid_optimizer.py:130
Methodfp32_param_copy_back_gpu_hook_closure
()
Megatron-LM/megatron/core/optimizer/cpu_offloading/hybrid_optimizer.py:129
Functionfp32_to_float16
Convert fp32 `val` to fp16/bf16
Megatron-LM/megatron/legacy/model/module.py:138
Functionfp8_cast_bf16
(path_src, path_dst)
slime/slime/utils/external_utils/command_utils.py:75
Functionfp8_dequantize
Dequantize a FP8 tensor to a higher precision.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:238
Functionfp8_pad_hook
FP8 requires class token length to be a multiple of 16 (for this model). Original model checkpoint may not be padded for FP8 so pad it here.
Megatron-LM/megatron/core/models/vision/radio.py:359
Methodfree
free a temporary bucket.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:511
Methodfree
free a temporary bucket.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:546
Methodfree
free a temporary bucket.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:629
Methodfree_up_grad_bucket
()
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:3217
Methodfreeze_patch_emb
(self)
Megatron-LM/megatron/legacy/model/vision/mit_backbone.py:293
Methodfreeze_pretrained_layers
(self, frozen_layers=[])
Megatron-LM/megatron/legacy/model/vision/esvit_swin_backbone.py:783
Methodfrom_config
Instantiate a `DynamicInferenceContext` from a `TransformerConfig` and an `InferenceWrapperConfig`.
Megatron-LM/megatron/core/inference/contexts/dynamic_context.py:665
Methodfrom_dict
(data: dict)
slime/slime/utils/types.py:79
Methodfrom_dict
(data: dict)
slime/slime/utils/types.py:110
Methodfrom_linear
(cls, linear, w_bit, group_size, init_only=False, scales=None, zeros=None)
slime/slime/backends/megatron_utils/megatron_to_hf/processors/quantizer_compressed_tensors.py:73
Methodfuse_layernorm_and_linear
TE backend chooses a single module for layernorm and linear
Megatron-LM/megatron/core/extensions/transformer_engine_spec_provider.py:45
Methodfuse_layernorm_and_linear
Does kitchen backend support a single module for layernorm and linear
Megatron-LM/megatron/core/extensions/kitchen.py:1044
Methodfuse_layernorm_and_linear
Does the backend support a single module for layernorm and linear
Megatron-LM/megatron/core/models/backends.py:53
Methodfuse_layernorm_and_linear
TE backend chooses a single module for layernorm and linear
Megatron-LM/megatron/core/models/backends.py:152
Functionfused_moe_backward_input_kernel
Backward kernel for computing grad_input. Forward: output = input @ weight.T (optionally multiplied by topk_weights) Backward: grad_inpu
slime/slime/backends/fsdp_utils/kernels/fused_moe_triton_backward_kernels.py:11
Functionfused_moe_backward_topk_weights_kernel
Backward kernel for computing grad_topk_weights. Forward: output = topk_weights * (input @ weight.T) Backward: grad_topk_weights = sum(g
slime/slime/backends/fsdp_utils/kernels/fused_moe_triton_backward_kernels.py:273
Functionfused_moe_backward_weight_kernel
Backward kernel for computing grad_weight. Forward: output = input @ weight.T (optionally multiplied by topk_weights) Backward: grad_wei
slime/slime/backends/fsdp_utils/kernels/fused_moe_triton_backward_kernels.py:132
Functionfuzzy_match
(result, rules)
gui-rl/desktop_env/evaluators/metrics/general.py:95
Functionfuzzy_place_math
(result_file_path, rules)
gui-rl/desktop_env/evaluators/metrics/general.py:101
Methodgather_at_indices
Run a PRM-teacher forward gathering log-probs at provided indices. Used by ``--distill-subset-mode student``. ``indices_per_sample`` is
slime/slime/backends/megatron_utils/actor.py:934
Functiongather_log_probs_at_indices
No-grad: gather GLOBAL log-probs at indices supplied via ``set_gather_at_indices``. Mirrors ``get_log_probs_and_entropy``'s signature so it slots
slime/slime/backends/megatron_utils/loss.py:424
Methodgen_file
(prefix="")
Megatron-LM/megatron/core/dist_checkpointing/strategies/filesystem_async.py:130
Functiongeneral_ict_model_provider
Build the model.
Megatron-LM/megatron/legacy/model/realm_model.py:18
Functiongenerate
(args, sample: Sample, sampling_params, evaluation: bool = False)
openclaw-rl/openclaw_api_server.py:168
Functiongenerate
(args, sample: Sample, sampling_params, evaluation: bool = False)
openclaw-rl/oel/openclaw_oel_api_server.py:241
Functiongenerate
( args, sample: Sample, sampling_params: Dict[str, Any], evaluation: bool = False, )
terminal-rl/generate.py:73
Functiongenerate
(args, sample: Sample, sampling_params, evaluation: bool = False)
openclaw-opd/openclaw_opd_api_server.py:232
Functiongenerate
(args, sample: Sample, sampling_params, evaluation: bool = False)
gui-rl/generate_with_gui.py:297
Functiongenerate
Offline generate: replay AgentNet screenshots through the policy model.
gui-rl/generate_with_agentnet.py:55
Functiongenerate
Custom generation function supporting tool calls. When ``evaluation=True`` (the dispatcher in ``slime.rollout.sglang_rollout`` detects the kw
toolcall-rl/generate_with_retool.py:579
Functiongenerate
(args, sample: Sample, sampling_params: dict)
swe-rl/generate_with_swe.py:25
Functiongenerate
Generate a complete agent-environment interaction trajectory for tau-bench. This is the main entry point for slime training. It creates a ta
slime/examples/tau-bench/generate_with_tau.py:100
Functiongenerate
Generate with TITO: tokens captured during generation, no retokenization.
slime/examples/strands_sglang/generate_with_strands.py:50
Functiongenerate
Custom multi-turn rollout that interacts with a pluggable environment.
slime/examples/geo3k_vlm_multi_turn/rollout.py:309
Functiongenerate
(args, sample: Sample, sampling_params)
slime/examples/search-r1/generate_with_search.py:145
Methodgenerate
Generates completions for a static list of prompts.
Megatron-LM/megatron/core/inference/engines/dynamic_engine.py:1260
Methodgenerate
( self, request: ChatInferenceRequest )
Megatron-LM/megatron/rl/inference/inference_interface.py:129
Methodgenerate
(self, rollout_id)
slime/slime/ray/rollout.py:148
Functiongenerate_rollout
Generate rollout for both training and evaluation.
slime/slime_plugins/rollout_buffer/rollout_buffer_example.py:305
Functiongenerate_rollout
An example to implement the generate_rollout function for an rule based rm rollout generation. Args: args: the whole args rollout
slime/slime/rollout/sft_rollout.py:17
Functiongenerate_rollout_openclaw
(args, rollout_id, data_buffer, evaluation=False)
openclaw-rl/openclaw_rollout.py:127
Functiongenerate_rollout_openclaw_combine
(args, rollout_id, data_buffer, evaluation=False)
openclaw-combine/openclaw_combine_rollout.py:131
Functiongenerate_rollout_openclaw_combine_select
``--rollout-function-path`` entry point for openclaw_combine_select. Same drain / eval logic as ``generate_rollout_openclaw_combine``; the on
openclaw-combine/openclaw_combine_select_rollout.py:153
Functiongenerate_rollout_openclaw_oel
Main rollout entry point for OpenClaw OEL. Supports all OEL modes: - online: continuous OPCD-style training (wait for external requests)
openclaw-rl/oel/openclaw_oel_rollout.py:141
Functiongenerate_rollout_openclaw_opd
(args, rollout_id, data_buffer, evaluation=False)
openclaw-opd/openclaw_opd_rollout.py:133
Functiongenerate_with_multi_agents
(args, sample: Sample, sampling_params, evaluation=False)
slime/examples/multi_agent/rollout_with_multi_agents.py:16
Functiongenerate_with_random_osl
(args: Namespace, sample: Sample, sampling_params: dict[str, Any])
slime/slime/rollout/generate_hub/benchmarkers.py:13
Methodgenerate_with_sglang
( self, *, args: Any, state: GenerateState, messages: List[Dict[str, A
gui-rl/agents/qwen3vl_agent.py:233
Methodget
(self, tag: str)
slime/slime/utils/tensor_backper.py:27
Methodget
(self, tag: str)
slime/slime/utils/tensor_backper.py:51
Methodget
(self, tag: str)
slime/slime/utils/tensor_backper.py:88
MethodgetCursorImageArray
(self)
gui-rl/desktop_env/server/pyxcursor.py:112
Functionget_a_and_b_segments
Divide sample into a and b segments.
Megatron-LM/megatron/legacy/data/dataset_utils.py:84
Functionget_accessibility_tree
()
gui-rl/desktop_env/server/main.py:903
Functionget_accessibility_tree
(env, *args)
gui-rl/desktop_env/evaluators/getters/misc.py:356
Methodget_active_request_count
(self)
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:44
Functionget_active_tab_html_parse
This function is used to get the specific element's text content from the active tab's html. config: Dict[str, str]{ # Ke
gui-rl/desktop_env/evaluators/getters/chrome.py:1487
Functionget_active_tab_info
This function is used to get all info about active tab. Warning! This function will reload the target-url page If the tartget url has cac
gui-rl/desktop_env/evaluators/getters/chrome.py:781
Functionget_active_tab_url_parse
This function is used to parse the url according to config["parse_keys"]. config: 'parse_keys': must exist, a list of ke
gui-rl/desktop_env/evaluators/getters/chrome.py:2258
Methodget_added_index
Get index that has been populated with vectors. Args: config (RetroPreprocessingConfig): Retro preprocessing config. Ret
Megatron-LM/megatron/core/datasets/retro/index/index.py:104
Functionget_advantages_and_returns
Function that computes advantages and returns from rewards and values. Calculated as in the original PPO paper: https://arxiv.org/abs/1707.06347
slime/slime/utils/ppo_utils.py:311
Functionget_all_ranks
Get caller's rank in tensor-model-parallel, data-parallel, context-parallel, pipeline-model-parallel and expert-model-parallel groups.
Megatron-LM/megatron/core/parallel_state.py:1997
Functionget_amax_reduction_group
Get the FP8 amax reduction group the caller rank belongs to.
Megatron-LM/megatron/core/parallel_state.py:1468
Functionget_annotated_datasets
(lines, threads=10)
Megatron-LM/examples/academic_paper_scripts/detxoify_lm/annotations/perspective_api_annotate.py:150
Functionget_audio_in_slide
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/impress.py:71
Functionget_background_image_in_slide
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/impress.py:10
Methodget_balanced_rollouts_if_remaining
(agent_id)
Megatron-LM/megatron/rl/agent/weighted_multi_task.py:214
Functionget_batch
(data_iterator)
Megatron-LM/tests/unit_tests/transformer/moe/test_upcycling.py:163
Methodget_batch_for_context_window
Returns the input data for inference This function gets called iteratively in the inference loop . It can be used to extract relevant input f
Megatron-LM/megatron/core/inference/model_inference_wrappers/abstract_model_inference_wrapper.py:141
Methodget_batch_for_context_window
Returns the inference data given context window This function gets called iteratively in a loop . Given the start and end context positions ,
Megatron-LM/megatron/core/inference/model_inference_wrappers/gpt/gpt_inference_wrapper.py:98
Methodget_batch_for_context_window
Returns the inference data given context window This function gets called iteratively in a loop . Given the start and end context positions ,
Megatron-LM/megatron/core/inference/model_inference_wrappers/multimodal/vlm_inference_wrapper.py:89
Functionget_bias_dropout_add_layer_scaling
Bias-dropout-add as in Megatron but with added LayerScaling handling.
Megatron-LM/examples/multimodal/layer_scaling.py:31
Methodget_block
Get the IDs for an evidence block plus the title of the corresponding document
Megatron-LM/megatron/legacy/data/ict_dataset.py:126
Functionget_block_samples_mapping
Get samples mapping for a dataset over fixed size blocks. This function also requires a dataset of the titles for the source documents since their
Megatron-LM/megatron/legacy/data/biencoder_dataset_utils.py:125
Functionget_bookmarks
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/chrome.py:465
Functionget_cache_file
Config: path (str): relative path in cache dir
gui-rl/desktop_env/evaluators/getters/file.py:159
Functionget_chrome_font_size
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/chrome.py:424
Functionget_chrome_language
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/chrome.py:390
Functionget_chrome_saved_address
(env, config: Dict[str, str])
gui-rl/desktop_env/evaluators/getters/chrome.py:985
← previousnext →7,601–7,700 of 10,561, ranked by callers