MCPcopy Create free account

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

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

↓ 1 callersFunctionimport_vm
Import the .ovf file into VirtualBox.
gui-rl/desktop_env/providers/virtualbox/manager.py:113
↓ 1 callersFunctioninfovqa_eval
Run InfoVQA evaluation.
Megatron-LM/examples/multimodal/evaluation/evaluate_infovqa.py:35
↓ 1 callersFunctioninit_cuda_graph_cache
Initialize the cache of modules for cuda graphs
Megatron-LM/megatron/core/transformer/utils.py:337
↓ 1 callersFunctioninit_distributed
(tp_size: int = 1, pp_size: int = 1)
Megatron-LM/examples/mimo/avlm_inference.py:23
↓ 1 callersFunctioninit_gloo_group
Initialize Gloo group for distributed communication.
slime/slime/utils/distributed_utils.py:20
↓ 1 callersFunctioninit_gpt_dataloader
( dp_group, micro_batch_size=1, vocab_size=50257, sequence_length=128, batch_size=8 )
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:179
↓ 1 callersFunctioninit_http_client
Initialize HTTP client and optionally enable distributed POST via Ray.
slime/slime/utils/http_utils.py:249
↓ 1 callersFunctioninit_hybrid_ep_buffer
Initialize the HybridEP buffer, including buffer allocation and metadata initialization. If a runtime dispatch/combine requires a larger
Megatron-LM/megatron/core/transformer/moe/fused_a2a.py:278
↓ 1 callersFunctioninit_indexed_dataset_infos
Gather meta-info about each indexed dataset. The returned info array allows for easy access to the configuration, and helps remove ambiguity.
Megatron-LM/megatron/core/datasets/retro/db/utils.py:38
↓ 1 callersMethodinit_model_chunk_offload_handler
Initialize a chunk offload handler for a model chunk (microbatch). Args: vp_size: Virtual pipeline size vp_s
Megatron-LM/megatron/core/pipeline_parallel/fine_grained_activation_offload.py:114
↓ 1 callersFunctioninit_model_parallel
(tp=1, pp=1, ep=1)
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:167
↓ 1 callersFunctioninit_persistent_async_worker
()
Megatron-LM/megatron/training/async_utils.py:21
↓ 1 callersFunctioninit_prm_engines
(args, pg, all_prm_engines)
slime/slime/ray/rollout.py:917
↓ 1 callersFunctioninit_proxy_pool
Initialize global proxy pool
gui-rl/desktop_env/providers/aws/proxy_pool.py:189
↓ 1 callersMethodinit_state_dict_from_bert
Initialize the state from a pretrained BERT model on iteration zero of ICT pretraining
Megatron-LM/megatron/legacy/model/realm_model.py:118
↓ 1 callersFunctioninitialize_distributed
Initialize torch.distributed and Megatron-Core model parallel groups. Args: tensor_model_parallel_size: Number of GPUs for tensor mo
Megatron-LM/examples/run_simple_mcore_train_loop.py:30
↓ 1 callersFunctioninitialize_distributed
(tensor_model_parallel_size=1, pipeline_model_parallel_size=1)
Megatron-LM/examples/export/trtllm_export/single_device_export/gpt_single_device_cpu_export.py:18
↓ 1 callersFunctioninitialize_distributed
(tensor_model_parallel_size=1, pipeline_model_parallel_size=1)
Megatron-LM/examples/export/trtllm_export/distributed_export/gpt_distributed_gpu_export.py:17
↓ 1 callersFunctioninitialize_gpt_model
( pre_process=True, post_process=True, seed=0, use_glu=True, **config_kwargs )
Megatron-LM/tests/unit_tests/dist_checkpointing/utils.py:27
↓ 1 callersMethodinitialize_megatron_env
Initialize Megatron global variables and fused kernels.
Megatron-LM/tools/checkpoint/loader_base.py:131
↓ 1 callersMethodinitialize_megatron_env
Initialize Megatron global variables and fused kernels.
Megatron-LM/tools/checkpoint/saver_base.py:136
↓ 1 callersFunctioninitialize_megatron_retro
Initialize megatron & save Retro config.
Megatron-LM/tools/retro/preprocess_data.py:57
↓ 1 callersFunctioninitialize_model_and_optimizer
Initialize model(s), optimizer, scheduler, and load from checkpoint. Args: args (Namespace): Runtime arguments. role (str): Logic
slime/slime/backends/megatron_utils/model.py:773
↓ 1 callersMethodinitialize_models
Construct a 3D(PPxEPxTP) array for models, fill it with None
Megatron-LM/tools/checkpoint/saver_base.py:295
↓ 1 callersMethodinitialize_registry
(self)
gui-rl/desktop_env/providers/azure/manager.py:26
↓ 1 callersMethodinitialize_registry
(self, **kwargs)
gui-rl/desktop_env/providers/aws/manager.py:199
↓ 1 callersMethodinitialize_registry
(self)
gui-rl/desktop_env/providers/virtualbox/manager.py:312
↓ 1 callersMethodinitialize_registry
(self, **kwargs)
gui-rl/desktop_env/providers/volcengine/manager.py:296
↓ 1 callersMethodinitialize_registry
(self)
gui-rl/desktop_env/providers/vmware/manager.py:300
↓ 1 callersMethodinitialize_registry
(self, **kwargs)
gui-rl/desktop_env/providers/aliyun/manager.py:279
↓ 1 callersFunctioninitialize_rng_tracker
Create the RNG tracker. 'use_te_rng_tracker' determines whether to use Megatron or TransformerEngine's implementation. In particular, Transfor
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:206
↓ 1 callersMethodinject_metadata
Return updated metadata merging overrides.
slime/slime/utils/eval_config.py:138
↓ 1 callersFunctioninprocess_restart
(train, args)
Megatron-LM/megatron/training/inprocess_restart.py:30
↓ 1 callersMethodinsert_megatron_path
(self)
Megatron-LM/tools/checkpoint/saver_hf_llava.py:37
↓ 1 callersMethodinsert_megatron_path_and_check_te
Check for an appropriate installation of transformer engine and add megatron to sys path.
Megatron-LM/tools/checkpoint/saver_base.py:38
↓ 1 callersMethodinsert_tensors
Reverse of `pop_tensors`. Replaces device type in sharded_tensors with actual values Value of `self` is considered to be the same aft
Megatron-LM/megatron/core/dist_checkpointing/tensor_aware_state_dict.py:213
↓ 1 callersMethodis_applicable_for_batch_dim
Checks if this batch dimension is applicable for the given real batch dimension. Applicable batch dimensions are those that have enou
Megatron-LM/megatron/core/inference/batch_dimensions_utils.py:40
↓ 1 callersMethodis_aux_loss_enabled
Check if the auxiliary loss is enabled.
Megatron-LM/megatron/core/transformer/moe/router.py:266
↓ 1 callersFunctionis_black
(rgb_str, threshold=50)
gui-rl/desktop_env/evaluators/metrics/slides.py:891
↓ 1 callersFunctionis_blockwise_float8tensor
Check if a tensor is a Blockwise FP8 tensor.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:162
↓ 1 callersFunctionis_blue
(rgb_str, threshold=50)
gui-rl/desktop_env/evaluators/metrics/slides.py:883
↓ 1 callersFunctionis_causal_conv1d_min_version
Check if minimum version of `causal_conv1d` is installed.
Megatron-LM/megatron/core/utils.py:457
↓ 1 callersFunctionis_child_of_bucket
(test_case, bucket)
Megatron-LM/tests/unit_tests/find_test_cases.py:27
↓ 1 callersFunctionis_color_blackish
(rgb_values, threshold=100)
gui-rl/desktop_env/evaluators/metrics/vlc.py:436
↓ 1 callersFunctionis_column_parallel_linear
Returns whether the given module is a ColumnParallelLinear layer.
Megatron-LM/megatron/core/fp8_utils.py:176
↓ 1 callersFunctionis_correct_minerva
Check if the solution is correct according to Minerva criteria. Args: solution_str: The solution string to check gt: The ground t
slime/slime/rollout/rm_hub/math_dapo_utils.py:185
↓ 1 callersFunctionis_correct_strict_box
Check if the prediction is correct using strict boxed answer criteria. Args: pred: The prediction string gt: The ground truth ans
slime/slime/rollout/rm_hub/math_dapo_utils.py:215
↓ 1 callersMethodis_current_async_call_done
Check if async save is finished on all ranks. For semantic correctness, requires rank synchronization in each check. This method must
Megatron-LM/megatron/core/dist_checkpointing/strategies/async_utils.py:133
↓ 1 callersFunctionis_dataloader_rank
Check if we should have the dataloader on this tensor and pipeline parallel rank.
Megatron-LM/examples/multimodal/dataloader_provider.py:84
↓ 1 callersFunctionis_deep_ep_available
()
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:1005
↓ 1 callersFunctionis_docker_available
Check if Docker is available and running.
swe-rl/mini-swe-agent/tests/environments/test_docker.py:10
↓ 1 callersMethodis_first_pp_rank
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_single_device_converter.py:134
↓ 1 callersFunctionis_flaky_failure
Assumes that certain keywords hint towards intermittent failures
Megatron-LM/tests/test_utils/python_scripts/launch_nemo_run_workload.py:18
↓ 1 callersFunctionis_green
(rgb_str, threshold=50)
gui-rl/desktop_env/evaluators/metrics/slides.py:887
↓ 1 callersFunctionis_hybrid_ep_available
()
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:1011
↓ 1 callersFunctionis_initialized
Useful for code segments that may be accessed with or without mpu initialization
Megatron-LM/megatron/core/parallel_state.py:1332
↓ 1 callersMethodis_kernel_available
Check whether the fused CUDA kernel can be used for the given shapes and settings. Args: mask (Optional[torch.Tensor]): Attention
Megatron-LM/megatron/core/fusions/fused_softmax.py:238
↓ 1 callersMethodis_last_pp_rank
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_single_device_converter.py:137
↓ 1 callersFunctionis_mamba_min_version
Check if minimum version of `mamba_ssm` is installed.
Megatron-LM/megatron/core/utils.py:425
↓ 1 callersFunctionis_mcore_tensor_parallel_duplicated
Check if the given parameter is Megatron-Core tensor model parallel and duplicated.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:725
↓ 1 callersFunctionis_mxfp8tensor
Check if a tensor is a Transformer Engine MXFP8Tensor
Megatron-LM/megatron/core/fp8_utils.py:108
↓ 1 callersFunctionis_pending
(target_branch: str)
Megatron-LM/tests/test_utils/python_scripts/check_status_of_main.py:46
↓ 1 callersFunctionis_pipeline_first_stage
Return True if in the first pipeline model-parallel stage, False otherwise.
Megatron-LM/megatron/core/parallel_state.py:1581
↓ 1 callersFunctionis_podman_available
Check if Podman is available and running.
swe-rl/mini-swe-agent/tests/environments/test_docker.py:19
↓ 1 callersFunctionis_red
(rgb_str, threshold=50)
gui-rl/desktop_env/evaluators/metrics/slides.py:879
↓ 1 callersFunctionis_red_color
(color)
gui-rl/desktop_env/evaluators/metrics/slides.py:100
↓ 1 callersFunctionis_retrieval_correct
(text: str, golden_answers: list[str])
slime/examples/search-r1/qa_em_format.py:147
↓ 1 callersFunctionis_row_parallel_linear
Returns whether the given module is a RowParallelLinear layer.
Megatron-LM/megatron/core/fp8_utils.py:188
↓ 1 callersFunctionis_swiglu_key
(key)
Megatron-LM/tools/checkpoint/checkpoint_inspector.py:459
↓ 1 callersFunctionis_te_min_version
Check if minimum version of `transformer-engine` is installed.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:146
↓ 1 callersMethodis_valid
Checks if the batch dimension is valid based on resource constraints. Args: max_requests: Maximum number of requests all
Megatron-LM/megatron/core/inference/batch_dimensions_utils.py:72
↓ 1 callersFunctionis_valid_file
(x: str)
Megatron-LM/megatron/legacy/data/image_folder.py:93
↓ 1 callersFunctionis_valid_sequence
(text)
slime/examples/search-r1/qa_em_format.py:51
↓ 1 callersFunctioniter_finalize_fn
()
Megatron-LM/megatron/training/checkpointing.py:647
↓ 1 callersMethoditer_worker_candidates
(self, start_idx: int)
terminal-rl/router_server.py:85
↓ 1 callersMethoditerate_train_actor
(self, iterator)
slime/slime/utils/profile_utils.py:41
↓ 1 callersMethoditerate_train_log_probs
(self, iterator)
slime/slime/utils/profile_utils.py:44
↓ 1 callersFunctionjsonl_reader
(p)
slime/slime/utils/data.py:34
↓ 1 callersMethodjudge_step
( self, args: Any, *, instruction: str, actions_history: list[str],
gui-rl/agents/qwen3vl_reward_agent.py:336
↓ 1 callersMethodjudge_step
( self, args: Any, *, problem_statement: str, step_debug: list[dict],
swe-rl/swe_prm.py:345
↓ 1 callersMethodjudge_turn
(self, turn_idx: int)
terminal-rl/rollout_agent.py:31
↓ 1 callersFunctionkarmarkar_karp
(seqlen_list: list[int], k_partitions: int, equal_size: bool)
slime/slime/utils/seqlen_balancing.py:20
↓ 1 callersMethodkernel
The forward kernel for the mainloop.
Megatron-LM/megatron/core/fusions/linear_cross_entropy/blackwell/fwd_mainloop.py:144
↓ 1 callersMethodkernel
The backward kernel for partial d_logits.
Megatron-LM/megatron/core/fusions/linear_cross_entropy/blackwell/bwd_partial_dlogits.py:130
↓ 1 callersFunctionkitchen_quantization_recipe_config
Loads a quantization recipe that uses a QAT_PARAMS recipe for all layers.
Megatron-LM/megatron/core/quantization/utils.py:31
↓ 1 callersFunctionknn_predict
(feature, feature_bank, feature_labels, classes, knn_k, knn_t)
Megatron-LM/megatron/legacy/model/vision/knn_monitor.py:105
↓ 1 callersFunctionlast_boxed_only_string
(string)
slime/slime/rollout/rm_hub/math_utils.py:379
↓ 1 callersMethodlaunch
(cls, env_cls: type[Agent], cls_args: dict, port: int, **kwargs)
Megatron-LM/megatron/rl/server/agent/fastapi_env_server.py:49
↓ 1 callersFunctionlaunch_and_wait_for_completion
( test_case: str, environment: str, n_repeat: int, time_limit: int, scope: str, contai
Megatron-LM/tests/test_utils/python_scripts/launch_jet_workload.py:45
↓ 1 callersFunctionlevenshtein_distance
(s1: str, s2: str)
Megatron-LM/examples/multimodal/evaluation/evaluate_vqav2.py:11
↓ 1 callersMethodlinear_fc1_forward_and_act
Do Linear FC1 and activation function forward. This function is used to overlap shared experts with the dispatcher. It is onl
Megatron-LM/megatron/core/transformer/moe/shared_experts.py:173
↓ 1 callersMethodlinear_fc2_forward
Do Linear FC2 forward. This function is used to overlap shared experts with the dispatcher. It is only useful when --moe-shar
Megatron-LM/megatron/core/transformer/moe/shared_experts.py:224
↓ 1 callersMethodlist_free_vms
(self, region)
gui-rl/desktop_env/providers/azure/manager.py:58
↓ 1 callersMethodload
Load the state of the data source
slime/slime/rollout/data_source.py:37
↓ 1 callersFunctionload_and_flatten
Wrapper function for doing all the fun at once.
Megatron-LM/tests/test_utils/python_scripts/recipe_parser.py:92
↓ 1 callersFunctionload_args_from_checkpoint
(args)
Megatron-LM/tools/checkpoint/loader_mixtral_hf.py:28
↓ 1 callersFunctionload_args_from_checkpoint
(args, model_size)
Megatron-LM/tools/checkpoint/loader_llama_mistral.py:280
↓ 1 callersMethodload_attributes
Load the necessary attributes: model, dataloader and empty BlockData
Megatron-LM/megatron/legacy/indexer.py:43
↓ 1 callersFunctionload_biencoder_checkpoint
selectively load retrieval models for indexing/retrieving from saved checkpoints
Megatron-LM/megatron/training/checkpointing.py:1868
↓ 1 callersFunctionload_checkpoint_no_arg_checks
(*args, **kwargs)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_layer_wise_optimizer.py:116
← previousnext →4,401–4,500 of 10,561, ranked by callers