MCPcopy Create free account

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

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

↓ 2 callersFunctionget_gpt_decoder_layer_specs
Helper function to get GPT block spec. Return a list of transformer layer spec of the current pipeline stage.
Megatron-LM/megatron/core/models/gpt/gpt_layer_specs.py:607
↓ 2 callersFunctionget_gpt_layer_with_inference_spec
Use this spec to use inference optimized linear layers. Args: qk_layernorm (bool, optional): To use layernorm for queries/keys. Defaults t
Megatron-LM/megatron/core/models/gpt/gpt_layer_specs.py:81
↓ 2 callersMethodget_grad
Get the grad of inputs
Megatron-LM/megatron/core/pipeline_parallel/utils.py:279
↓ 2 callersFunctionget_group_world_size_rank
()
Megatron-LM/pretrain_ict.py:36
↓ 2 callersFunctionget_hierarchical_context_parallel_groups
Get the inner ring of context parallel group the caller rank belongs to.
Megatron-LM/megatron/core/parallel_state.py:1435
↓ 2 callersFunctionget_host_info
()
slime/slime/utils/http_utils.py:42
↓ 2 callersMethodget_index
(self, bound, fps, max_frame, first_idx=0)
Megatron-LM/examples/multimodal/evaluation/evaluation_datasets.py:1248
↓ 2 callersFunctionget_index_cache_path
Get the index cache path for the given path Args: idx_path (str): The path to the index file object_storage_config (ObjectStorag
Megatron-LM/megatron/core/datasets/object_storage_utils.py:170
↓ 2 callersFunctionget_individual_db_dir
Individual DB's directory. Args: project_dir (str): Path to Retro project dir. prefix (str): Unique relative path to dataset with
Megatron-LM/megatron/core/datasets/retro/db/utils.py:148
↓ 2 callersFunctionget_inference_context
The inference context manages the KV cache and other inference state.
Megatron-LM/examples/inference/gpt/gpt_dynamic_inference.py:139
↓ 2 callersFunctionget_inference_controller
Buid text generation controller, which manages the model inference context. Args: model (MegatronModule): Megatron GPT model. con
Megatron-LM/examples/inference/gpt/gpt_dynamic_inference.py:195
↓ 2 callersMethodget_label_date
Get most recent date when label was attached.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder_github.py:120
↓ 2 callersFunctionget_language_model_config
(config)
Megatron-LM/examples/multimodal/config.py:9
↓ 2 callersFunctionget_language_model_spec
Get the language model spec.
Megatron-LM/tests/unit_tests/models/test_mimo_model.py:120
↓ 2 callersMethodget_layer
Gets the transformer layer at the specified index.
Megatron-LM/megatron/core/models/common/model_chunk_schedule_plan.py:406
↓ 2 callersFunctionget_layer_spec
(is_vit, normalization)
Megatron-LM/examples/multimodal/layer_specs.py:48
↓ 2 callersFunctionget_layer_spec_te
(is_vit=False, padding=False)
Megatron-LM/examples/multimodal/layer_specs.py:96
↓ 2 callersFunctionget_llava_projection_config
Return a TransformerConfig for the vision projection MLP.
Megatron-LM/examples/mimo/configs/llava_avlm.py:25
↓ 2 callersFunctionget_llava_projection_layer_spec
Layer spec for the vision-projection MLP.
Megatron-LM/examples/mimo/configs/llava_avlm.py:50
↓ 2 callersMethodget_local_model_weights_per_gpu
Get the trtllm model weights split per gpu Given the trtllm mapping information (tp, pp rank etc) we split the model weights in a list, with
Megatron-LM/megatron/core/export/trtllm/trtllm_weights_converter/single_device_trtllm_model_weights_converter.py:422
↓ 2 callersFunctionget_logprob_and_entropy
Compute log probabilities and entropy. Parameters: logits: Model output logits with shape [seq_len, vocab_size] target_tokens: Ta
slime/slime/backends/fsdp_utils/actor.py:909
↓ 2 callersMethodget_loss_mask_with_multimodal_alignment
( self, messages: list[dict], input_ids: list[int], tools: list[dict] = None )
slime/slime/utils/mask_utils.py:142
↓ 2 callersFunctionget_mamba_stack_modelopt_spec
Mix the native spec with TENorm. This is essentially the native local spec except for the layernorm implementation is using TENorm from Trans
Megatron-LM/megatron/core/post_training/modelopt/mamba/model_specs.py:19
↓ 2 callersFunctionget_mamba_version
Get mamba version from __version__; if not available use pip's. Use caching.
Megatron-LM/megatron/core/utils.py:404
↓ 2 callersFunctionget_mcore_tensor_parallel_partition_dim
Get the partition dimension for a Megatron-Core tensor model parallel parameter.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:732
↓ 2 callersFunctionget_mem_size_str
Convert number of bytes to human-readable string.
Megatron-LM/megatron/core/inference/contexts/dynamic_context.py:180
↓ 2 callersFunctionget_memory_usage
Get current memory usage in MB
toolcall-rl/tool_sandbox.py:41
↓ 2 callersFunctionget_merged_db_path_map
Paths to merged datasets. Args: project_dir (str): Path to Retro project dir. Returns: A dict of chunk databases, one for ea
Megatron-LM/megatron/core/datasets/retro/db/utils.py:243
↓ 2 callersFunctionget_mesh_names
Get all the sub-mesh names in the DeviceMesh.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:95
↓ 2 callersMethodget_micro_batch_size
Get current global batch size.
Megatron-LM/megatron/core/num_microbatches_calculator.py:283
↓ 2 callersFunctionget_microbatch_id_in_model_chunk
Helper method to get the microbatch_id within model chunk given the iteration number.
Megatron-LM/megatron/core/pipeline_parallel/schedules.py:1106
↓ 2 callersFunctionget_minimum_num_micro_batch_size
(total_lengths, max_tokens_per_gpu)
slime/slime/utils/data.py:256
↓ 2 callersMethodget_mismatch_errors
Return list of detailed errors for mismatched cudagraph args.
Megatron-LM/megatron/core/transformer/cuda_graphs.py:920
↓ 2 callersMethodget_mixer
(self, tp_size=1, cp_size=1, use_mem_eff_path=True)
Megatron-LM/tests/unit_tests/ssm/test_mamba_mixer.py:24
↓ 2 callersFunctionget_mlp_module_spec
(use_te: bool = True)
Megatron-LM/examples/multimodal/layer_specs.py:182
↓ 2 callersFunctionget_mlp_module_spec
(use_te: bool = True)
Megatron-LM/examples/multimodal/nvlm/internvit.py:157
↓ 2 callersMethodget_mode
Method to get the operating mode
Megatron-LM/megatron/core/rerun_state_machine.py:247
↓ 2 callersFunctionget_model
Initialize model and load checkpoint.
Megatron-LM/examples/inference/gpt/gpt_dynamic_inference.py:101
↓ 2 callersFunctionget_model_and_buffers
( input_dim: int, output_dim: int, num_layers: int, bias: bool, shared_embedding: bool,
Megatron-LM/tests/unit_tests/distributed/test_param_and_grad_buffer.py:15
↓ 2 callersMethodget_model_cls
(self)
slime/slime/backends/fsdp_utils/actor.py:186
↓ 2 callersFunctionget_model_provider_func
( args: argparse.Namespace, role: Literal["actor", "critic"] = "actor", )
slime/slime/backends/megatron_utils/model_provider.py:57
↓ 2 callersFunctionget_models
(count, dtype)
Megatron-LM/tools/checkpoint/loader_legacy.py:136
↓ 2 callersFunctionget_models
(count, dtype, pre_process, post_process)
Megatron-LM/tools/checkpoint/saver_legacy.py:208
↓ 2 callersMethodget_modules
(cls, config, use_transformer_engine, use_gpu)
Megatron-LM/tests/unit_tests/transformer/test_retro_attention.py:40
↓ 2 callersFunctionget_moe_module_spec_for_backend
Helper function to get module spec for MoE
Megatron-LM/megatron/core/models/gpt/moe_module_specs.py:32
↓ 2 callersFunctionget_mtbench_chat_data
Return a MTBench dataset.
Megatron-LM/megatron/post_training/utils.py:27
↓ 2 callersFunctionget_mtp_layer_spec_for_backend
Get the MTP layer spec. Returns: ModuleSpec: Module specification with modules from the backend.
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:427
↓ 2 callersFunctionget_new_p2pop_function
(func)
slime/slime/utils/reloadable_process_group.py:90
↓ 2 callersMethodget_next
Return the next micro-batch for the requested keys. - If `micro_batch_indices` is provided, selects rows according to the current i
slime/slime/backends/megatron_utils/data.py:281
↓ 2 callersFunctionget_norm_mlp_module_spec_te
()
Megatron-LM/examples/multimodal/layer_specs.py:193
↓ 2 callersFunctionget_num_chunks_per_sample
Compute seq_length // chunk_length. Args: sample_length (int): Alias of `sequence_length`. chunk_length (int): Retro chunk length
Megatron-LM/megatron/core/datasets/retro/utils.py:79
↓ 2 callersFunctionget_output_path
Get output path for a given test. Args: num_cuda_graphs (int): Number of cuda graphs used. trial_idx (int): Trial index, for when
Megatron-LM/tests/functional_tests/test_cases/gpt/gpt_dynamic_inference_tp1_pp1_583m_cuda_graphs_validation/cuda_graphs.py:22
↓ 2 callersFunctionget_output_path
Generation output path.
Megatron-LM/examples/multimodal/run_text_generation.py:424
↓ 2 callersFunctionget_padding
Calculate padding needed for SP, CP, TP comm overlap, and FP8. Args: seq_len (int): Model sequence length. cp_size (int): Context
Megatron-LM/megatron/core/models/multimodal/context_parallel.py:9
↓ 2 callersFunctionget_pairs
Return set of symbol pairs in a word. Word is represented as tuple of symbols (symbols being variable-length strings).
Megatron-LM/megatron/training/tokenizer/gpt2_tokenization.py:79
↓ 2 callersFunctionget_paragraph_texts_odt
(document)
gui-rl/desktop_env/evaluators/metrics/docs.py:170
↓ 2 callersMethodget_paused_used
Compute number of paused blocks used.
Megatron-LM/megatron/core/inference/contexts/dynamic_block_allocator.py:57
↓ 2 callersFunctionget_pp_rank_microbatches
Get the number of total, warmup, and remaining microbatches in PP scheduling.
Megatron-LM/megatron/core/pipeline_parallel/schedules.py:730
↓ 2 callersMethodget_prompt
Return a tuple with the prompt string and the golden data.
Megatron-LM/megatron/rl/agent/reward_only_agent.py:52
↓ 2 callersMethodget_pt_num_samples_and_chunks
Number of samples & chunks (e.g., 32*n_samples) in corpus.
Megatron-LM/tools/retro/cli/cli.py:145
↓ 2 callersMethodget_pt_sample
(cls, data_key: str, idx: int)
Megatron-LM/tools/retro/cli/cli.py:172
↓ 2 callersFunctionget_quant_config_or_none
Resolve quantization config for a layer.
Megatron-LM/megatron/core/quantization/utils.py:9
↓ 2 callersMethodget_quantization_context
Return appropriate quantization context (FP8 or FP4) in cudagraph mode.
Megatron-LM/megatron/core/transformer/cuda_graphs.py:637
↓ 2 callersFunctionget_qwen35_rope_index
( spatial_merge_size: int, image_token_id: int, video_token_id: int, vision_start_token_id: in
slime/slime_plugins/megatron_bridge/qwen35_vl.py:515
↓ 2 callersFunctionget_reformulation_metadata
Reads MCore data for N-D flattened tensors from checkpoint metadata during ckpt load. Args: sharded_state_dict (ShardedStateDict): sharde
Megatron-LM/megatron/core/dist_checkpointing/strategies/torch.py:865
↓ 2 callersMethodget_relative_seq_len
Function to get the rotary sequence length. Args: inference_context (BaseInferenceContext): Used during Inference time
Megatron-LM/megatron/core/models/common/embeddings/relative_pos_embedding.py:138
↓ 2 callersFunctionget_response_lengths
(loss_masks: list[list[int]])
slime/slime/utils/mask_utils.py:4
↓ 2 callersMethodget_response_lengths
(self, loss_masks: list[list[int]])
slime/slime/utils/mask_utils.py:15
↓ 2 callersFunctionget_retro_decoder_block_spec
Retro decoder block spec. Retro decoder block implementation details: - The retro decoder block consists of interleaved GPT layers an
Megatron-LM/megatron/core/models/retro/decoder_spec.py:126
↓ 2 callersFunctionget_rgb_from_color
(color)
gui-rl/desktop_env/evaluators/metrics/slides.py:771
↓ 2 callersMethodget_root_mesh
Get the device mesh.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:615
↓ 2 callersFunctionget_sb_environment
(config: dict, instance: dict)
swe-rl/mini-swe-agent/src/minisweagent/run/extra/swebench.py:80
↓ 2 callersFunctionget_schedule_table
Get the schedule table for PP scheduling.
Megatron-LM/megatron/core/pipeline_parallel/schedules.py:790
↓ 2 callersMethodget_scores
Get attribute scores for a given text via Perspective API. :param input_text: the input text :param requested_attributes: the
Megatron-LM/examples/academic_paper_scripts/detxoify_lm/perspective_api.py:44
↓ 2 callersMethodget_screenshot
Gets a screenshot from the server. With the cursor. None -> no screenshot or unexpected error.
gui-rl/desktop_env/controllers/python.py:42
↓ 2 callersMethodget_sharded_state_dict
()
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fully_parallel.py:106
↓ 2 callersFunctionget_slack_user_id
Look up Slack user ID by email with caching and retries.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder.py:212
↓ 2 callersMethodget_slack_user_id
Get Slack user ID from email.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder_github.py:104
↓ 2 callersFunctionget_slide_background_color
(slide)
gui-rl/desktop_env/evaluators/metrics/slides.py:253
↓ 2 callersFunctionget_slime_extra_args_provider
(add_custom_arguments=None)
slime/slime/utils/arguments.py:35
↓ 2 callersFunctionget_split_dim
(tensor_name)
Megatron-LM/tools/checkpoint/hybrid_conversion.py:45
↓ 2 callersMethodget_stage
Get current review stage.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder_github.py:137
↓ 2 callersMethodget_start_end_idx
Calculate the start and end indices for the current data parallel worker's chunk within a batch. Returns: tuple:
Megatron-LM/megatron/training/datasets/data_samplers.py:137
↓ 2 callersMethodget_start_end_idx_global_batch
(self)
Megatron-LM/megatron/training/datasets/data_samplers.py:182
↓ 2 callersMethodget_stats
Get cache statistics.
slime/slime/router/middleware_hub/radix_tree.py:524
↓ 2 callersFunctionget_swin
(is_teacher=False)
Megatron-LM/megatron/legacy/model/vision/esvit_swin_backbone.py:810
↓ 2 callersMethodget_system_prompt
( self, processed_width: Optional[int] = None, processed_height: Optional[int] = None,
gui-rl/agents/qwen35_agent.py:219
↓ 2 callersMethodget_system_prompt
( self, processed_width: Optional[int] = None, processed_height: Optional[int] = None,
gui-rl/agents/qwen3vl_agent.py:180
↓ 2 callersMethodget_template_vars
(self)
swe-rl/mini-swe-agent/src/minisweagent/models/litellm_model.py:82
↓ 2 callersFunctionget_tensor_and_context_parallel_group
Get the tensor- and context-parallel group the caller rank belongs to.
Megatron-LM/megatron/core/parallel_state.py:1510
↓ 2 callersFunctionget_tensor_model_parallel_src_rank
Calculate the global rank corresponding to the first local rank in the tensor model parallel group.
Megatron-LM/megatron/core/parallel_state.py:1649
↓ 2 callersFunctionget_tensor_shapes
Determine right tensor sizes (based on position of rank with respect to split rank) and model size.
Megatron-LM/megatron/core/pipeline_parallel/schedules.py:1963
↓ 2 callersFunctionget_tensor_shapes_for_tp
(transformer_config, tp_size)
Megatron-LM/tests/unit_tests/transformer/test_transformer_layer.py:298
↓ 2 callersMethodget_terminal_output
Gets the terminal output from the server. None -> no terminal output or unexpected error.
gui-rl/desktop_env/controllers/python.py:92
↓ 2 callersFunctionget_test_data
Load test fixtures from a trajectory JSON file
swe-rl/mini-swe-agent/tests/conftest.py:30
↓ 2 callersFunctionget_test_vocab_size
(make_divisible_by=128)
Megatron-LM/tests/unit_tests/dist_checkpointing/models/common.py:181
↓ 2 callersFunctionget_text_dataset_for_adding
Convert GPT token chunk dataset to a text dataset for passing to the embedder. Args: config (RetroPreprocessingConfig): Retro preproc
Megatron-LM/megatron/core/datasets/retro/index/build.py:257
↓ 2 callersFunctionget_text_dataset_for_training
Convert GPT token chunk dataset to a text dataset for passing to the embedder. Args: config (RetroPreprocessingConfig): Retro preproc
Megatron-LM/megatron/core/datasets/retro/index/build.py:141
↓ 2 callersMethodget_text_embeddings
Get embeddings for text tokens in the input. Args: input_ids: Input token IDs of shape [batch_size, seq_len] containing text token
Megatron-LM/megatron/core/models/mimo/model/base.py:181
↓ 2 callersMethodget_text_from_loss_mask
(self, token_ids: list[int], loss_masks: list[int])
slime/slime/utils/mask_utils.py:169
← previousnext →2,001–2,100 of 10,561, ranked by callers