MCPcopy Create free account

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

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

↓ 1 callersMethodget_layer_static_inputs
Get the static inputs for the layer. We assume that the module has one hidden_states input, whose shape is inferred from the
Megatron-LM/megatron/core/transformer/module.py:188
↓ 1 callersFunctionget_llava_projection_config
Return a TransformerConfig for the vision projection MLP.
Megatron-LM/examples/mimo/configs/llava_vlm.py:82
↓ 1 callersFunctionget_llava_projection_layer_spec
Layer spec for the vision-projection MLP.
Megatron-LM/examples/mimo/configs/llava_vlm.py:107
↓ 1 callersFunctionget_load_checkpoint_path_by_args
Get the checkpoint path based on the arguments.
Megatron-LM/megatron/training/checkpointing.py:183
↓ 1 callersFunctionget_local_gpu_id
()
slime/slime/ray/train_actor.py:20
↓ 1 callersMethodget_logical_hybrid_fsdp_rank
Returns the logical rank of the current process within the full-shard hybrid FSDP group. In full-shard hybrid FSDP, parameters are f
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/utils.py:621
↓ 1 callersMethodget_loss_scale
Get current loss scale factor. NOTE: The output should be a CUDA tensor of size 1.
Megatron-LM/megatron/core/optimizer/optimizer.py:239
↓ 1 callersMethodget_lr
Compute the learning rates for each parameter group. Returns: list[float]: A list of learning rates, one for each parameter group
slime/slime/backends/fsdp_utils/lr_scheduler.py:148
↓ 1 callersFunctionget_lr_scheduler
Create and configure the learning-rate scheduler. This configures iteration-based schedules derived from the global batch size and run-time a
slime/slime/backends/fsdp_utils/lr_scheduler.py:157
↓ 1 callersFunctionget_ltor_masks_and_position_ids
Build masks and position id for left to right model.
Megatron-LM/examples/multimodal/train.py:152
↓ 1 callersFunctionget_macys_product_url_parse
Parse Macy's product url path, extract: - mens_clothing: true if 'mens-clothing' in path, else None - shirts: true if any key 'Top_style'
gui-rl/desktop_env/evaluators/getters/chrome.py:2334
↓ 1 callersFunctionget_mamba_layer_spec_te
(padding=False)
Megatron-LM/examples/multimodal/layer_specs.py:125
↓ 1 callersFunctionget_mamba_version_str
()
Megatron-LM/megatron/core/utils.py:411
↓ 1 callersMethodget_mask
Create a mask for the specified tokens based on the given order. Args: order (str): The order of parallelism types (e.g., 'tp-dp-
Megatron-LM/megatron/core/parallel_state.py:486
↓ 1 callersFunctionget_mask_start_and_end_idx
Returns a list of tuples holding the start and end index in arr of the non-zeros contiguuous sub arrays. For instance, if arr = [0, 1, 0
Megatron-LM/examples/multimodal/train.py:168
↓ 1 callersFunctionget_master_placeholder_color
(prs)
gui-rl/desktop_env/evaluators/metrics/slides.py:105
↓ 1 callersMethodget_max_sequence_lengths
Maximum sequence length for active requests.
Megatron-LM/megatron/core/inference/contexts/dynamic_context.py:767
↓ 1 callersFunctionget_mcore_reviewers
Get all members of mcore-reviewers group and its subgroups recursively.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder.py:108
↓ 1 callersFunctionget_megatron_model_type
(model_name: str)
slime/examples/geo3k_vlm_multi_turn/run_geo3k_vlm_multi_turn.py:26
↓ 1 callersFunctionget_megatron_optimizer_config
Return a Megatron optimizer config object from Megatron's arguments.
Megatron-LM/megatron/training/training.py:1181
↓ 1 callersMethodget_mem_alloc_context
Get the memory allocation context for the parameter and gradient buffers.
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:1664
↓ 1 callersFunctionget_merged_sampled_dataset
Get sampled dataset (for training the vector index). Args: project_dir (str): Path to Retro project dir. chunk_length (int): GPT
Megatron-LM/megatron/core/datasets/retro/db/utils.py:308
↓ 1 callersFunctionget_merged_state_dict
Return a merged (base + LoRA) state dict on CPU. Merges the adapter in-place, copies the state dict, then unmerges. The returned dict uses ba
slime/slime/backends/fsdp_utils/lora_utils.py:156
↓ 1 callersFunctionget_merged_train_dataset
Get training dataset (for adding to the vector index). Args: project_dir (str): Path to Retro project dir. chunk_length (int): GP
Megatron-LM/megatron/core/datasets/retro/db/utils.py:332
↓ 1 callersFunctionget_mlp_module_spec
(use_te: bool = True)
Megatron-LM/examples/multimodal/radio/radio_g.py:45
↓ 1 callersFunctionget_mock_language_layer_spec
Get the mock layer specification for the language model. Returns: ModuleSpec: Mock specification for language model layers
Megatron-LM/examples/mimo/configs/mock.py:97
↓ 1 callersFunctionget_mock_language_model_config
Create a mock language model configuration. Args: config: Optional base configuration to modify Returns: TransformerCon
Megatron-LM/examples/mimo/configs/mock.py:28
↓ 1 callersFunctionget_mock_projection_config
Create a mock projection layer configuration. Args: hidden_size: Hidden dimension size (used as the vision projection output size)
Megatron-LM/examples/mimo/configs/mock.py:76
↓ 1 callersFunctionget_mock_projection_layer_spec
Get the mock layer specification for the projection layer. Returns: ModuleSpec: Mock specification for projection layers
Megatron-LM/examples/mimo/configs/mock.py:120
↓ 1 callersFunctionget_mock_vision_layer_spec
Get the mock layer specification for the vision model. Args: normalization: Type of normalization to use Returns: Modul
Megatron-LM/examples/mimo/configs/mock.py:107
↓ 1 callersFunctionget_mock_vision_model_config
Create a mock vision model configuration. Args: config: Optional base configuration to modify Returns: TransformerConfi
Megatron-LM/examples/mimo/configs/mock.py:47
↓ 1 callersFunctionget_mock_vlm_dataloader
Create a DataLoader for mock VLM data. Args: batch_size: Batch size dataset_size: Size of the dataset image_size: Si
Megatron-LM/examples/mimo/data/mock.py:165
↓ 1 callersMethodget_model_argv
Get argv list for customizing initialization.
Megatron-LM/tests/functional_tests/test_cases/common/ckpt_converter/__main__.py:146
↓ 1 callersMethodget_model_argv
LLaVA model args.
Megatron-LM/tests/functional_tests/test_cases/common/ckpt_converter/__main__.py:615
↓ 1 callersFunctionget_model_provider
(only_query_model=False, only_context_model=False, biencoder_shared_query_context_model=False)
Megatron-LM/megatron/legacy/model/biencoder_model.py:19
↓ 1 callersFunctionget_model_size_str
(model)
Megatron-LM/examples/inference/gpt/utils.py:362
↓ 1 callersFunctionget_modelopt_torch_quantization_config
Return a quantization config.
Megatron-LM/examples/post_training/modelopt/quantize.py:100
↓ 1 callersFunctionget_moe_model_and_buffers
( num_layers: int, hidden_size: int, num_moe_experts: int, moe_grouped_gemm: bool, ep_size
Megatron-LM/tests/unit_tests/distributed/test_grad_sync_with_expert_parallel.py:56
↓ 1 callersFunctionget_moe_module_spec
Helper function to get module spec for MoE
Megatron-LM/megatron/core/models/gpt/moe_module_specs.py:13
↓ 1 callersFunctionget_mtp_num_layers_to_build
Get the number of MTP layers to build.
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:527
↓ 1 callersFunctionget_mtp_ranks
Get the ranks of the MTP layers.
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:494
↓ 1 callersFunctionget_named_params
(args, state_dict)
slime/tools/convert_torch_dist_to_hf.py:100
↓ 1 callersFunctionget_neighbor_dir
Get directory containing neighbor IDs for a dataset (i.e., train, valid, or test). Args: project_dir (str): Retro project dir. ke
Megatron-LM/megatron/core/datasets/retro/query/utils.py:22
↓ 1 callersMethodget_neighbor_tokens
(cls, sample_id: int, chunk_id: int, data_key: str="train")
Megatron-LM/tools/retro/cli/cli.py:176
↓ 1 callersMethodget_new_request_id
Gets a new request id
Megatron-LM/megatron/core/inference/scheduler.py:37
↓ 1 callersMethodget_next_data_parallel_rank
Selects the next data parallel rank using round-robin scheduling. Returns: bytes: The ZMQ identity of the next data para
Megatron-LM/megatron/core/inference/data_parallel_inference_coordinator.py:120
↓ 1 callersMethodget_next_proxy
Get next available proxy
gui-rl/desktop_env/providers/aws/proxy_pool.py:64
↓ 1 callersFunctionget_non_matching_values
(tensor1_shard, tensor2_shard)
Megatron-LM/tests/unit_tests/distributed/test_reduce_scatter_with_fp32_accumulation.py:16
↓ 1 callersFunctionget_norm_mlp_module_spec_te
()
Megatron-LM/examples/multimodal/radio/radio_g.py:56
↓ 1 callersMethodget_num_stages_from_str
Get the number of PP * VPP stages from a layout string.
Megatron-LM/megatron/core/transformer/pipeline_parallel_layer_layout.py:277
↓ 1 callersMethodget_num_unfinalized_calls
Get the number of active async calls.
Megatron-LM/megatron/core/dist_checkpointing/strategies/async_utils.py:589
↓ 1 callersMethodget_number_of_tokens_per_expert
Get the number of tokens per expert.
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:1231
↓ 1 callersMethodget_object
Get the file from S3
Megatron-LM/megatron/core/datasets/object_storage_utils.py:61
↓ 1 callersFunctionget_object_storage_access
Get the object storage access
Megatron-LM/megatron/core/datasets/object_storage_utils.py:215
↓ 1 callersMethodget_obs
(self, lease_id: str)
gui-rl/env_client.py:45
↓ 1 callersMethodget_obs
(self, lease_id: str)
gui-rl/env_pool_server.py:223
↓ 1 callersMethodget_observation
Execute the action and return the observation.
swe-rl/mini-swe-agent/src/minisweagent/agents/default.py:100
↓ 1 callersMethodget_offload_event
Get the CUDA event for a named offload operation.
Megatron-LM/megatron/core/pipeline_parallel/fine_grained_activation_offload.py:346
↓ 1 callersFunctionget_one_epoch_dataloader
Specifically one epoch to be used in an indexing job.
Megatron-LM/megatron/legacy/data/biencoder_dataset_utils.py:28
↓ 1 callersFunctionget_open_retrieval_batch
(data_iterator)
Megatron-LM/megatron/legacy/data/orqa_wiki_dataset.py:28
↓ 1 callersFunctionget_open_retrieval_wiki_dataset
()
Megatron-LM/megatron/legacy/data/orqa_wiki_dataset.py:16
↓ 1 callersFunctionget_optim_param_to_id_map
Generate mapping from optimizer param to optimizer state id.
Megatron-LM/megatron/core/dist_checkpointing/optimizer.py:35
↓ 1 callersFunctionget_optimizer_param_scheduler
Build the learning rate scheduler.
Megatron-LM/megatron/training/training.py:1125
↓ 1 callersFunctionget_optimizer_param_scheduler
Create and configure the optimizer learning-rate/weight-decay scheduler. This configures iteration-based schedules derived from the global batch
slime/slime/backends/megatron_utils/model.py:36
↓ 1 callersMethodget_output
Gets the module forward output. This function is used to overlap shared experts with the dispatcher. It is only useful when -
Megatron-LM/megatron/core/transformer/moe/shared_experts.py:259
↓ 1 callersMethodget_packed_batch
Create a packed sequence batch with multiple sequences of varying lengths. Args: seq_lengths: List of sequence lengths (
Megatron-LM/tests/unit_tests/transformer/test_multi_token_prediction.py:251
↓ 1 callersMethodget_padded_vocab_size
Return the paded vocab size We extract the lm head and vocab embedding and use that to determine padded_vocab_size Returns:
Megatron-LM/megatron/core/export/trtllm/trtllm_weights_converter/single_device_trtllm_model_weights_converter.py:405
↓ 1 callersMethodget_parameter_state_dp_reshardable
Get internal representation of parameter state without any copies and modifications. This is referred to as "fully sharded bucket space" beca
Megatron-LM/megatron/core/optimizer/distrib_optimizer.py:933
↓ 1 callersFunctionget_params
(model)
Megatron-LM/examples/post_training/modelopt/prune.py:150
↓ 1 callersMethodget_partitions
(self)
slime/slime/utils/seqlen_balancing.py:55
↓ 1 callersFunctionget_path_count
keeps tracks of number of times we've seen the input `path` and return count-1
Megatron-LM/megatron/core/config_logger.py:40
↓ 1 callersFunctionget_path_with_count
calls get_path_count and appends returned value to path
Megatron-LM/megatron/core/config_logger.py:52
↓ 1 callersMethodget_permuted_hidden_states_by_experts
(self, hidden_states: torch.Tensor)
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:1285
↓ 1 callersMethodget_pg_collection
(self)
Megatron-LM/tests/unit_tests/ssm/test_mamba_block.py:26
↓ 1 callersFunctionget_policy_first_response
Get the policy model's first response to a question. Uses turn_type="side" so no training data is generated. For Student: just ask the m
openclaw-rl/oel/eval/gsm8k_personal_agent.py:217
↓ 1 callersMethodget_pos_enc
Get positional encoding for certain input size
Megatron-LM/megatron/core/models/vision/radio.py:257
↓ 1 callersFunctionget_pp_last_rank
Return the global rank of the last rank in the pipeline parallel group.
Megatron-LM/megatron/core/pipeline_parallel/utils.py:51
↓ 1 callersFunctionget_priority
Get priority based on age category with custom Slack emojis.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder.py:201
↓ 1 callersFunctionget_prompt_and_generated
Strip prompt and other unnecessary text from generation.
Megatron-LM/examples/multimodal/run_text_generation.py:648
↓ 1 callersMethodget_queue_size
(self)
openclaw-rl/openclaw_rollout.py:85
↓ 1 callersMethodget_queue_size
(self)
openclaw-rl/oel/openclaw_oel_rollout.py:98
↓ 1 callersMethodget_queue_size
(self)
openclaw-opd/openclaw_opd_rollout.py:89
↓ 1 callersMethodget_queue_size
(self)
openclaw-combine/openclaw_combine_rollout.py:89
↓ 1 callersMethodget_queue_size
(self)
openclaw-combine/openclaw_combine_select_rollout.py:105
↓ 1 callersFunctionget_radio_g_layer_spec
(normalization)
Megatron-LM/examples/multimodal/radio/radio_g.py:65
↓ 1 callersFunctionget_radio_g_layer_spec_te
()
Megatron-LM/examples/multimodal/radio/radio_g.py:105
↓ 1 callersFunctionget_random_prompt_tokens
(tokenizer, num_input_tokens)
Megatron-LM/tools/run_inference_performance_test.py:196
↓ 1 callersMethodget_ready_bucket_group_for_reduction
Checks if all buckets in the bucket group containing the given bucket_id are ready for gradient reduction. If so, returns the list of
Megatron-LM/megatron/core/distributed/fsdp/src/megatron_fsdp/param_and_grad_buffer.py:3050
↓ 1 callersFunctionget_recent_milestones
Get the two most recent milestones from the project.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder.py:91
↓ 1 callersFunctionget_reinforce_plus_plus_baseline_advantages
Calculates the unwhitened advantages for the REINFORCE++-baseline algorithm. Broadcasting the scalar (reward - group_baseline) to each token.
slime/slime/utils/ppo_utils.py:281
↓ 1 callersFunctionget_reinforce_plus_plus_returns
Calculates discounted returns for REINFORCE++ (https://arxiv.org/pdf/2501.03262) Args: rewards (Tensor): A tensor of scalar rewards
slime/slime/utils/ppo_utils.py:211
↓ 1 callersMethodget_reload_event
Get the CUDA event for a named reload operation.
Megatron-LM/megatron/core/pipeline_parallel/fine_grained_activation_offload.py:350
↓ 1 callersFunctionget_replay
(env, trajectory: List[Dict[str, Any]])
gui-rl/desktop_env/evaluators/getters/replay.py:4
↓ 1 callersFunctionget_requests_from_file
Get requests from a file.
Megatron-LM/examples/inference/gpt/utils.py:305
↓ 1 callersFunctionget_required_reviewers
Get list of required reviewers who haven't approved yet.
Megatron-LM/tests/test_utils/python_scripts/auto_reminder.py:150
↓ 1 callersMethodget_restored_hidden_states_by_experts
(self, hidden_states: torch.Tensor)
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:1312
↓ 1 callersFunctionget_retriever
(config)
slime/examples/search-r1/local_dense_retriever/retrieval_server.py:272
↓ 1 callersFunctionget_retro_datasets
Get train, valid, test retro datasets. Args: config (RetroConfig): Retro preprocessing config. gpt_datasets (dict): Mapping of da
Megatron-LM/megatron/core/datasets/retro/query/retro_dataset.py:158
↓ 1 callersFunctionget_retro_encoder_block_spec
Retro encoder block spec. The retro encoder block consists of one customized Retro encoder layer (layer 1), and all of the following layers a
Megatron-LM/megatron/core/models/retro/encoder_spec.py:121
↓ 1 callersFunctionget_retro_preprocessing_config
()
Megatron-LM/tools/retro/preprocess_data.py:199
← previousnext →4,201–4,300 of 10,561, ranked by callers