MCPcopy Create free account

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

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

Functionis_unexpected_key
(x: ShardedBase)
Megatron-LM/megatron/core/dist_checkpointing/validation.py:275
Methodis_unexpectedly_large
Helper method to estimate whether a result is unexpectedly large. Some calculation errors manifest themselves as results with unexpectedly la
Megatron-LM/megatron/core/rerun_state_machine.py:649
Functionis_unitialized
Check if parallel state has been initialized Deprecated. Use is_initialized instead.
Megatron-LM/megatron/core/parallel_state.py:1337
Functionis_utc_0
Format as: Local time: Thu 2024-01-25 12:56:06 WET Universal time: Thu 2024-01-25 12:56:06 UTC RTC time: Thu 2024
gui-rl/desktop_env/evaluators/metrics/basic_os.py:17
Functionis_valid_group
(group, min_valid_group_size, task_type="math")
slime/slime_plugins/rollout_buffer/generator/base_generator.py:330
Functionis_vlc_fullscreen
(actual_window_size, screen_size)
gui-rl/desktop_env/evaluators/metrics/vlc.py:173
Functionis_vlc_playing
Checks if VLC is currently playing a file.
gui-rl/desktop_env/evaluators/metrics/vlc.py:21
Functionis_vlc_recordings_folder
Checks if VLC's recording folder is set to the expected value.
gui-rl/desktop_env/evaluators/metrics/vlc.py:140
Functionjoin_str_list
Join a list of strings, handling spaces appropriately
Megatron-LM/megatron/legacy/data/realm_dataset_utils.py:65
Functionjoin_str_list
Join a list of strings, handling spaces appropriately
Megatron-LM/megatron/legacy/data/biencoder_dataset_utils.py:78
Methodjudge_turn
(self, turn_idx: int)
terminal-rl/agent/prm_agent.py:214
Methodkill
(self)
Megatron-LM/megatron/rl/inference/megatron.py:249
Methodkill
(self)
Megatron-LM/megatron/rl/server/inference/inference_interface_server.py:85
Methodkill
(self)
Megatron-LM/megatron/rl/server/agent/fastapi_env_server.py:101
Functionla_freq_type
Frequency between LA (linear attention) layers and SDPA (scaled dot-product attention) layers. Accepts either: - An integer N: Represents a (
Megatron-LM/megatron/training/arguments.py:337
Methodlaunch
(cls, model: GPTModel, **kwargs)
Megatron-LM/megatron/rl/inference/megatron.py:212
Methodlaunch
(cls, interface_cls: type[InferenceInterface], **kwargs)
Megatron-LM/megatron/rl/server/inference/inference_interface_server.py:50
Functionlaunch_app
()
gui-rl/desktop_env/server/main.py:240
Methodlayer_norm
Which module to use for layer norm
Megatron-LM/megatron/core/extensions/transformer_engine_spec_provider.py:53
Methodlayer_norm
Which module to use for layer norm
Megatron-LM/megatron/core/extensions/kitchen.py:1059
Methodlayer_norm
Which module for layernorm
Megatron-LM/megatron/core/models/backends.py:63
Methodlinear
Which linear module TE backend uses
Megatron-LM/megatron/core/extensions/transformer_engine_spec_provider.py:33
Methodlist_free_vms
List the paths of VM that are free to use allocated.
gui-rl/desktop_env/providers/base.py:79
Methodlist_free_vms
(self, region=DEFAULT_REGION, lock_needed=True, **kwargs)
gui-rl/desktop_env/providers/aws/manager.py:226
Methodlist_free_vms
(self)
gui-rl/desktop_env/providers/docker/manager.py:109
Methodlist_free_vms
(self, lock_needed=True)
gui-rl/desktop_env/providers/virtualbox/manager.py:419
Methodlist_free_vms
(self, lock_needed=True, **kwargs)
gui-rl/desktop_env/providers/volcengine/manager.py:323
Methodlist_free_vms
(self, lock_needed=True)
gui-rl/desktop_env/providers/vmware/manager.py:407
Methodlist_free_vms
(self, lock_needed=True, **kwargs)
gui-rl/desktop_env/providers/aliyun/manager.py:306
Functionlist_images
()
swe-rl/server/swe_exec_server.py:96
Methodlist_workers
List all registered workers
slime/slime/router/router.py:199
Functionliteral_match
(result: Any, expected: Any, **options)
gui-rl/desktop_env/evaluators/metrics/general.py:59
Functionllava_avlm_dataloader_provider
(train_val_test_num_samples)
Megatron-LM/examples/mimo/data/energon_avlm_task_encoder.py:357
Functionllava_embedding_ranks
LLaVA's embedding ranks consist of the first and last ranks of the pipeline. Args: pp_ranks: A list of global ranks that constitute a pipe
Megatron-LM/pretrain_vlm.py:408
Functionllava_embedding_ranks
LLava's embedding ranks consist of the decoder's first and last ranks (ie, the ViT has no embeddings). Args: pp_ranks: A list of global ra
Megatron-LM/examples/multimodal/train.py:300
Functionllava_position_embedding_ranks
LLaVA's positional embeddings are on the first rank stage Args: pp_ranks: A list of global ranks that constitute a pipeline group.
Megatron-LM/pretrain_vlm.py:422
Functionllava_position_embedding_ranks
LLava's embedding ranks consist of the singular rank of the model or the decoder's first rank. Args: pp_ranks: A list of global ranks that
Megatron-LM/examples/multimodal/train.py:313
Functionllava_vlm_dataloader_provider
(train_val_test_num_samples, is_video_input=False)
Megatron-LM/examples/mimo/data/energon_vlm_task_encoder.py:279
Methodload
(self, rollout_id=None)
gui-rl/gui_data_source.py:267
Methodload
Main load method.
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:110
Methodload
Translates MCore ShardedTensors to PyT ShardedTensors & loads from PyT Distributed fmt. Args: sharded_state_dict (ShardedStateDic
Megatron-LM/megatron/core/dist_checkpointing/strategies/torch.py:914
Methodload
(self, sharded_state_dict: ShardedStateDict, checkpoint_dir: Union[str, Path])
Megatron-LM/megatron/core/dist_checkpointing/strategies/tensorstore.py:47
Methodload
(self, sharded_state_dict: ShardedStateDict, checkpoint_dir: Union[str, Path])
Megatron-LM/megatron/core/dist_checkpointing/strategies/zarr.py:207
Methodload
(self, rollout_id=None)
slime/slime/rollout/data_source.py:132
Functionload_and_upcycle_model
Load a dense model checkpoint and convert it to a MoE model. This function loads a checkpoint for a dense model and converts it to the MoE m
Megatron-LM/megatron/core/transformer/moe/upcycling_utils.py:329
Functionload_checkpoint
Required top-level function that creates the loader, calls its .load(), and handles exceptions by signaling 'exit'.
Megatron-LM/tools/checkpoint/loader_core.py:82
Functionload_checkpoint
(queue, args)
Megatron-LM/tools/checkpoint/loader_mixtral_hf.py:340
Functionload_checkpoint
(queue, args)
Megatron-LM/tools/checkpoint/loader_legacy.py:384
Functionload_checkpoint
Required top-level function that creates the loader, calls its .load(), and handles exceptions by signaling 'exit'.
Megatron-LM/tools/checkpoint/loader_llava.py:352
Functionload_checkpoint
(queue, args)
Megatron-LM/tools/checkpoint/loader_llama_mistral.py:663
Methodload_common
Load common (non-sharded) objects state dict from the checkpoint. Args: checkpoint_dir (Union[str, Path]): checkpoint directory
Megatron-LM/megatron/core/dist_checkpointing/strategies/common.py:74
Functionload_distributed_checkpoint
(checkpoint_path, gpt_model)
Megatron-LM/examples/export/trtllm_export/single_device_export/gpt_single_device_cpu_export.py:50
Functionload_distributed_checkpoint
(checkpoint_path, gpt_model)
Megatron-LM/examples/export/trtllm_export/distributed_export/gpt_distributed_gpu_export.py:49
Methodload_parameter_state
Load the distributed parameter state from disk. Args: filename (str): path to load parameter state from.
Megatron-LM/megatron/core/optimizer/distrib_optimizer.py:2237
Methodload_parameter_state
(self, *args, **kwargs)
Megatron-LM/tests/unit_tests/test_checkpointing.py:69
Methodload_rand
(x)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fully_parallel.py:76
Functionload_script
(config_path: str)
Megatron-LM/tests/test_utils/python_scripts/generate_local_jobs.py:19
Methodload_sharded_base
(x: Any)
Megatron-LM/megatron/core/dist_checkpointing/tensor_aware_state_dict.py:311
Methodload_sharded_metadata
(self, checkpoint_dir: Path)
Megatron-LM/megatron/core/dist_checkpointing/strategies/fully_parallel.py:398
Methodload_sharded_metadata
Load sharded metadata from the checkpoint for ShardedTensors and ShardedObjects. Returns a dictionary similar to a sharded state dict, but no
Megatron-LM/megatron/core/dist_checkpointing/strategies/base.py:158
Methodload_sharded_metadata
Uses tensors and objects metadata stored in the metadata file.
Megatron-LM/megatron/core/dist_checkpointing/strategies/torch.py:1020
Methodload_sharded_metadata
(self, checkpoint_dir: Union[str, Path])
Megatron-LM/megatron/core/dist_checkpointing/strategies/common.py:153
Methodload_sharded_object
(sh_obj: ShardedObject)
Megatron-LM/megatron/core/dist_checkpointing/strategies/common.py:115
Methodload_sharded_objects
Replaces all ShardedObject from a given state dict with values loaded from the checkpoint. Args: sharded_objects_state_di
Megatron-LM/megatron/core/dist_checkpointing/strategies/common.py:100
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/classification.py:89
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/realm_model.py:198
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/gpt_model.py:117
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/language_model.py:600
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/biencoder_model.py:320
Methodload_state_dict
Customize load.
Megatron-LM/megatron/legacy/model/transformer.py:1804
Methodload_state_dict
Customize load.
Megatron-LM/megatron/legacy/model/bert_model.py:84
Methodload_state_dict
Customized load.
Megatron-LM/megatron/legacy/model/multiple_choice.py:100
Methodload_state_dict
Method to restore the state saved as a serializable dict.
Megatron-LM/megatron/core/rerun_state_machine.py:1088
Methodload_state_dict
Method to restore the state saved as a serializable dict.
Megatron-LM/megatron/core/rerun_state_machine.py:1270
Methodload_state_dict
No-op because tensors are already loaded in-place by `_load_base_checkpoint` with FSDP2.
Megatron-LM/megatron/core/distributed/torch_fully_sharded_data_parallel.py:150
Methodload_state_dict
Load the state dictionary into the module.
Megatron-LM/megatron/core/distributed/fsdp/mcore_fsdp_adapter.py:133
Methodload_state_dict
Load pass-in `state_dict`.
Megatron-LM/megatron/core/optimizer/optimizer.py:273
Methodload_state_dict
(self, state_dict)
Megatron-LM/megatron/core/optimizer/optimizer.py:849
Methodload_state_dict
(self, state_dict)
Megatron-LM/megatron/core/optimizer/optimizer.py:1216
Methodload_state_dict
(self, state_dict: Dict)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:34
Methodload_state_dict
(self, state_dict)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:49
Methodload_state_dict
(self, state_dict: Dict)
Megatron-LM/megatron/core/optimizer/grad_scaler.py:139
Methodload_state_dict
(self, state_dict)
Megatron-LM/tests/unit_tests/test_checkpointing.py:63
Methodload_state_dict
(self, state_dict)
slime/slime/backends/fsdp_utils/checkpoint.py:43
Methodload_state_dict
(self, state_dict)
slime/slime/backends/fsdp_utils/checkpoint.py:58
Methodload_tensors_metadata
(self, checkpoint_dir: Path)
Megatron-LM/megatron/core/dist_checkpointing/strategies/fully_parallel.py:395
Methodload_tensors_metadata
(self, checkpoint_dir: Path)
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:263
Methodload_tensors_metadata
(self, checkpoint_dir: Union[str, Path])
Megatron-LM/megatron/core/dist_checkpointing/strategies/tensorstore.py:63
Methodload_tensors_metadata
(self, checkpoint_dir: Union[str, Path])
Megatron-LM/megatron/core/dist_checkpointing/strategies/zarr.py:216
Methodload_tensors_metadata
(self, checkpoint_dir: Path)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fully_parallel.py:88
Methodload_weights_hf_to_megatron
Store hf_pretrained before calling parent's load method.
slime/slime_plugins/megatron_bridge/qwen3_5_text.py:37
Methodload_with_flag
(strict)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:864
Functionlocal_multi_tensor_applier
Multi tensor op applier
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:105
Functionlocal_multi_tensor_l2_norm
Computes l2 norm for a list of contiguous tensors works as a drop-in replacement for amp_C.multi_tensor_l2norm
Megatron-LM/megatron/core/utils.py:1164
Functionlocal_multi_tensor_scale
Works as a drop-in replacement for amp_C.multi_tensor_scale.
Megatron-LM/megatron/core/utils.py:1176
Functionlocal_multi_tensor_scale
Works as a drop-in replacement for amp_C.multi_tensor_scale.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/mixed_precision.py:109
Functionlocal_test_data
Load local test fixtures
swe-rl/mini-swe-agent/tests/conftest.py:60
Functionlocal_test_specs
()
Megatron-LM/tests/unit_tests/test_tokenizer.py:26
Functionlog
(message)
Megatron-LM/megatron/rl/logging.py:18
Functionlog
(args, metrics, step_key: str)
slime/slime/utils/logging_utils.py:35
← previousnext →8,101–8,200 of 10,561, ranked by callers