MCPcopy Create free account

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

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

Methodvocab
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:843
Methodvocab
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:916
Methodvocab
Vocab.
Megatron-LM/megatron/training/tokenizer/multimodal_tokenizer.py:328
Methodvocab
Vocab.
Megatron-LM/megatron/training/tokenizer/sft_tokenizer.py:184
Methodvocab_file
Returns vocabulary file path if specified.
Megatron-LM/megatron/core/tokenizers/text/text_tokenizer.py:246
Methodvocab_size
Returns tokenizer vocab size.
Megatron-LM/megatron/core/tokenizers/base_tokenizer.py:41
Methodvocab_size
Returns vocabulary size.
Megatron-LM/megatron/core/tokenizers/text/text_tokenizer.py:150
Methodvocab_size
Returns vocab size.
Megatron-LM/megatron/core/tokenizers/text/libraries/null_tokenizer.py:50
Methodvocab_size
Returns size of tokenizer vocabulary.
Megatron-LM/megatron/core/tokenizers/text/libraries/huggingface_tokenizer.py:274
Methodvocab_size
Returns tokenizer vocab size.
Megatron-LM/megatron/core/tokenizers/text/libraries/tiktoken_tokenizer.py:296
Methodvocab_size
The vocabulary size
Megatron-LM/megatron/core/datasets/megatron_tokenizer.py:97
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:146
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:237
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:342
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:442
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:806
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:912
Methodvocab_size
Vocabulary size.
Megatron-LM/megatron/training/tokenizer/multimodal_tokenizer.py:338
Methodvocab_size
Vocabulary size.
Megatron-LM/megatron/training/tokenizer/sft_tokenizer.py:194
Methodvocab_size
(self)
Megatron-LM/megatron/training/tokenizer/bert_tokenization.py:213
Methodwake_up
(self, tags)
slime/slime/ray/train_actor.py:108
Methodweight
(name: str)
Megatron-LM/megatron/core/pipeline_parallel/multimodule_communicator.py:501
Functionweight_dequant_kernel
(x_ptr, s_ptr, y_ptr, M, N, BLOCK_SIZE: tl.constexpr)
slime/tools/fp8_cast_bf16.py:15
Methodwindow_select
(pos_embed)
Megatron-LM/megatron/core/models/vision/radio.py:290
Functionwith_defer
(deferred_func)
slime/slime/utils/timer.py:92
Methodwithout_data
(self)
Megatron-LM/megatron/core/dist_checkpointing/mapping.py:250
Methodwithout_data
(self)
Megatron-LM/megatron/core/dist_checkpointing/mapping.py:572
Methodwithout_data
(self)
Megatron-LM/megatron/core/dist_checkpointing/mapping.py:658
Functionworker_init_fn
(_)
Megatron-LM/megatron/training/datasets/data_samplers.py:75
Functionworker_process
(task_queue, done_queue, rollout_func, reward_func, client, sampling_params)
slime/slime_plugins/rollout_buffer/generator/base_generator.py:108
Methodworker_thread_func
(self)
openclaw-rl/openclaw_rollout.py:49
Methodworker_thread_func
(self)
openclaw-rl/oel/openclaw_oel_rollout.py:62
Methodworker_thread_func
(self)
openclaw-opd/openclaw_opd_rollout.py:53
Methodworker_thread_func
(self)
openclaw-combine/openclaw_combine_rollout.py:53
Methodworker_thread_func
(self)
openclaw-combine/openclaw_combine_select_rollout.py:67
Methodworld_size
Can be called to get configured world of this instance Returns: int: World size configured for this instance
Megatron-LM/megatron/core/utils.py:1875
Functionwrap_non_main_replicas
(x: Any)
Megatron-LM/megatron/core/dist_checkpointing/strategies/fully_parallel.py:489
Functionwrapped
(*args, **kwargs)
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:42
Functionwrapped
(**kwargs)
slime/slime/utils/external_utils/typer_utils.py:30
Functionwrapped_func
(*args, **kwarg)
Megatron-LM/megatron/core/models/gpt/fine_grained_callables.py:37
Functionwrapped_func
Check if we should skip executing the original function based on the current step condition and the tensor store status. If t
Megatron-LM/megatron/core/transformer/moe/moe_utils.py:1222
Functionwrapped_local_to_global
(*args, **kwargs)
slime/tools/convert_hf_to_torch_dist.py:37
Functionwrapped_map_mcore_to_hf
(mcore_weights_name)
slime/tools/convert_hf_to_torch_dist.py:80
Functionwrapped_model_provider
(pre_process, post_process, add_encoder=True, add_decoder=True)
Megatron-LM/examples/multimodal/run_text_generation.py:849
Functionwrapped_model_provider
(pre_process, post_process)
Megatron-LM/examples/multimodal/model_converter/vision_model_tester.py:55
Functionwrapped_model_provider
( pre_process: bool = True, post_process: bool = True, vp_stage: int | None = None )
slime/slime/backends/megatron_utils/model_provider.py:64
Functionwrapped_provider
(pre_process=True, post_process=True, vp_stage=None, **extra_kwargs)
slime/slime/backends/megatron_utils/model_provider.py:300
Functionwrapped_weight_to_mcore_format
(mcore_weights_name, hf_weights)
slime/tools/convert_hf_to_torch_dist.py:83
Functionwrapper
(*args, **kwargs)
Megatron-LM/megatron/core/utils.py:2321
Functionwrapper
(*args, **kwargs)
slime/slime/utils/timer.py:76
Methodwrite_data
Write all items from ``plan``.
Megatron-LM/megatron/core/dist_checkpointing/strategies/filesystem_async.py:420
Functionwrite_data_os_err_mock_fn
Raises an error on worker #2 during storage save
Megatron-LM/tests/unit_tests/dist_checkpointing/test_async_save.py:17
Functionwrite_online_eval_to_tensorboard
Write online evaluation data to Tensorboard.
Megatron-LM/examples/multimodal/train.py:363
Methodwrite_preloaded_data
Performs actual data saving to storage. Args: local_proc_idx (int): index of a local process that performs writing
Megatron-LM/megatron/core/dist_checkpointing/strategies/filesystem_async.py:342
Methodwrite_preloaded_data_multiproc
Performs saving data to storage with multiple processes. Starts predefined number of processes and uses 2 queues to make sure the re
Megatron-LM/megatron/core/dist_checkpointing/strategies/filesystem_async.py:235
Functionwrite_to_buffer
(request: Request)
slime/slime_plugins/rollout_buffer/buffer.py:260
Methodzero_grad
Zero gradients and prepare for next forward pass.
Megatron-LM/megatron/core/optimizer/optimizer.py:234
Methodzero_grad
We only need to zero the model related parameters, i.e., float16_groups & fp32_from_fp32_groups. We additionally zero fp32_from_float1
Megatron-LM/megatron/core/optimizer/optimizer.py:710
Methodzero_grad
(self, set_to_none=True)
Megatron-LM/megatron/core/optimizer/optimizer.py:1133
Methodzero_grad_buffer
Zeros out all grad buffers. Needs to be called at the beginning of each training iteration.
Megatron-LM/megatron/core/distributed/distributed_data_parallel.py:539
Methodzero_parameters
Zero out all parameters in embedding.
Megatron-LM/megatron/core/models/common/embeddings/language_model_embedding.py:88
← previous10,501–10,561 of 10,561, ranked by callers