MCPcopy Create free account

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

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

Methodsetup_method
(self, method, use_cpu_initialization=False, swiglu=True)
Megatron-LM/tests/unit_tests/transformer/moe/test_grouped_mlp.py:29
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_moe_layer_discrepancy.py:21
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_shared_experts.py:15
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_shared_experts.py:77
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_aux_loss.py:72
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_aux_loss.py:136
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_aux_loss.py:181
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_multihot_indices_converter.py:37
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/transformer/moe/test_sequential_mlp.py:74
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/ssm/test_mamba_layer.py:17
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/ssm/test_mamba_block.py:22
Methodsetup_method
(self, tp_size, sp, cp_size)
Megatron-LM/tests/unit_tests/ssm/test_gated_delta_net.py:54
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/ssm/test_mamba_mixer.py:18
Methodsetup_method
Setup method
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_distributed_gpu_converter.py:28
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/export/trtllm/test_single_device_fp8.py:189
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/export/trtllm/test_distributed_fp8.py:192
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_layer_1f1b.py:250
Methodsetup_method
(self, method)
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_chunk_1f1b.py:68
Functionsetup_moe_params
Setup MOE parameters for testing.
slime/tests/test_fused_experts_backward.py:272
Functionsft_loss_function
Compute supervised fine-tuning loss over response tokens. Computes log-probabilities of the ground-truth tokens in the response segments and
slime/slime/backends/megatron_utils/loss.py:1136
Functionsh_ten_build_fn
(*args, **kwargs)
Megatron-LM/megatron/core/dist_checkpointing/strategies/resharding.py:274
Functionsh_ten_build_fn
( key: str, t: torch.Tensor, replica_id: ReplicaId, flattened_range: Optional[slice] )
Megatron-LM/megatron/core/transformer/mlp.py:281
Functionsh_ten_build_fn
( key: str, t: torch.Tensor, replica_id: ReplicaId, flattened_range: Optional[slice] )
Megatron-LM/megatron/core/ssm/gated_delta_net.py:537
Functionsh_ten_build_fn
( key: str, t: torch.Tensor, replica_id: ReplicaId, flattened_range: Optional[slice] )
Megatron-LM/megatron/core/ssm/mamba_mixer.py:1166
Methodsh_ten_build_fn
( key: str, t: torch.Tensor, replica_id: ReplicaId, flattened_
Megatron-LM/megatron/core/transformer/moe/experts.py:354
Functionsh_ten_merge_fn
(sub_state_dict)
Megatron-LM/megatron/core/dist_checkpointing/strategies/resharding.py:279
Functionsh_ten_merge_fn
(sub_state_dict)
Megatron-LM/megatron/core/transformer/mlp.py:382
Functionsh_ten_merge_fn
(sub_state_dict)
Megatron-LM/megatron/core/ssm/gated_delta_net.py:569
Functionsh_ten_merge_fn
(sub_state_dict)
Megatron-LM/megatron/core/ssm/mamba_mixer.py:1198
Methodsh_ten_merge_fn
(sub_state_dict, tp_axis: int, with_glu: bool)
Megatron-LM/megatron/core/transformer/moe/experts.py:626
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/post_training/modelopt/layers.py:151
Methodsharded_state_dict
Non-default implementation for embeddings due to `allow_shape_mismatch` param
Megatron-LM/megatron/core/tensor_parallel/layers.py:311
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/tensor_parallel/layers.py:1069
Methodsharded_state_dict
Sharding along axis 1, bias not sharded
Megatron-LM/megatron/core/tensor_parallel/layers.py:1320
Methodsharded_state_dict
Replicate cross TP/DP.
Megatron-LM/megatron/core/extensions/kitchen.py:362
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/extensions/kitchen.py:426
Methodsharded_state_dict
Sharding along axis 1, bias not sharded
Megatron-LM/megatron/core/extensions/kitchen.py:489
Methodsharded_state_dict
For each gemm, sharding along axis 0, bias sharded. Assume sharded_offsets[-1] is the expert parallel offset.
Megatron-LM/megatron/core/extensions/kitchen.py:802
Methodsharded_state_dict
For each gemm, sharding along axis 1, bias not sharded. Assume sharded_offsets[-1] is the expert parallel offset.
Megatron-LM/megatron/core/extensions/kitchen.py:855
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/extensions/kitchen.py:1009
Methodsharded_state_dict
Replicate cross TP/DP.
Megatron-LM/megatron/core/extensions/transformer_engine.py:441
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/extensions/transformer_engine.py:637
Methodsharded_state_dict
Sharding along axis 0, bias sharded
Megatron-LM/megatron/core/extensions/transformer_engine.py:737
Methodsharded_state_dict
Sharding along axis 1, bias not sharded
Megatron-LM/megatron/core/extensions/transformer_engine.py:837
Methodsharded_state_dict
Sharded state dict for the learnable softmax offset parameter
Megatron-LM/megatron/core/extensions/transformer_engine.py:1143
Methodsharded_state_dict
For each gemm, sharding along axis 0, bias sharded. Assume sharded_offsets[-1] is the expert parallel offset.
Megatron-LM/megatron/core/extensions/transformer_engine.py:1604
Methodsharded_state_dict
For each gemm, sharding along axis 1, bias not sharded. Assume sharded_offsets[-1] is the expert parallel offset.
Megatron-LM/megatron/core/extensions/transformer_engine.py:1650
Methodsharded_state_dict
Sharded state dict implementation handling duplication of encoder and decoder layers. Some layers (output, embedding) are shared between the
Megatron-LM/megatron/core/models/T5/t5_model.py:476
Methodsharded_state_dict
Sharded state dict implementation for GPTModel backward-compatibility. Removing extra state. Tie word embeddings and output layer in
Megatron-LM/megatron/core/models/gpt/gpt_model.py:784
Methodsharded_state_dict
Sharded state dict implementation that handles the output layer weights tying. Args: prefix (str): Module name prefix.
Megatron-LM/megatron/core/models/common/language_module/language_module.py:340
Methodsharded_state_dict
Get sharded state dict. Args: prefix (str): Module name prefix. sharded_offsets (tuple): Offsets of local shard withi
Megatron-LM/megatron/core/models/retro/model.py:92
Methodsharded_state_dict
Generate a sharded state dictionary for the transformer block. Args: prefix (str, optional): Prefix to be added to all k
Megatron-LM/megatron/core/transformer/transformer_block.py:774
Methodsharded_state_dict
Generate a sharded state dictionary for the multi token prediction module. Args: prefix (str, optional): Prefix to be ad
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:1189
Methodsharded_state_dict
Default implementation for sharded state dict for distributed checkpointing. General definition of sharded_state_dict simply calls `sharded_s
Megatron-LM/megatron/core/transformer/module.py:57
Methodsharded_state_dict
Retrieve sharded_state_dict from the module being wrapped.
Megatron-LM/megatron/core/transformer/module.py:474
Methodsharded_state_dict
Sharded state dict for the learnable softmax offset parameter
Megatron-LM/megatron/core/transformer/dot_product_attention.py:272
Methodsharded_state_dict
Gets sharded state dict.
Megatron-LM/megatron/core/transformer/moe/shared_experts.py:132
Methodsharded_state_dict
Maps local expert to global experts. The sharded state dict is interchangable with SequentialMLP's.
Megatron-LM/megatron/core/transformer/moe/experts.py:989
Methodsharded_state_dict
Maps local expert to global experts.
Megatron-LM/megatron/core/transformer/moe/experts.py:1157
Methodsharded_state_dict
Provide a sharded state dictionary for distributed checkpointing.
Megatron-LM/megatron/core/ssm/gated_delta_net.py:428
Methodsharded_state_dict
Returns a sharded state dictionary for the current object. This function constructs a sharded state dictionary by iterating over the
Megatron-LM/megatron/core/ssm/mamba_block.py:316
Methodsharded_state_dict
Sharding along axis 0, bias not sharded
Megatron-LM/megatron/core/ssm/mamba_mixer.py:82
Methodsharded_state_dict
Provide a sharded state dictionary for distributed checkpointing.
Megatron-LM/megatron/core/ssm/mamba_mixer.py:1057
Methodsharded_state_dict
Generate a sharded state dictionary for the mamba layer. Args: prefix (str, optional): Prefix to be added to all keys in
Megatron-LM/megatron/core/ssm/mamba_layer.py:136
Methodsharded_state_dict
Builds sharded state dict for the optimizer, based on model's sharded state dict. Args: model_sharded_state_dict (ShardedStateDic
Megatron-LM/megatron/core/optimizer/optimizer.py:307
Methodsharded_state_dict
( self, model_sharded_state_dict: ShardedStateDict, is_loading: bool = False,
Megatron-LM/megatron/core/optimizer/optimizer.py:800
Methodsharded_state_dict
( self, model_sharded_state_dict: ShardedStateDict, is_loading: bool = False, **kwargs )
Megatron-LM/megatron/core/optimizer/optimizer.py:1184
Methodsharded_state_dict
Chooses between 3 param state sharding implementations as requested by `metadata['distrib_optim_sharding_type']`. Sharding t
Megatron-LM/megatron/core/optimizer/distrib_optimizer.py:1166
Methodsharded_state_dict
Sharded state dict for torch_dist format checkpointing. For fixed DP usage only, set replica_id to 0 for all ShardedTensor.
Megatron-LM/megatron/core/optimizer/layer_wise_optimizer.py:248
Methodsharded_state_dict
(self, *args, metadata: Optional[dict] = None, **kwargs)
Megatron-LM/tests/unit_tests/test_checkpointing.py:72
Methodsharded_state_dict
(self)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_optimizer.py:91
Methodsharded_state_dict
(self)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_optimizer.py:171
Methodshared_embedding_or_output_weight
(self)
slime/slime_plugins/megatron_bridge/qwen35_vl.py:678
Functionshutdown
Shutdowns fault folerance, updates the FT timeouts if possible
Megatron-LM/megatron/training/ft_integration.py:205
Methodshutdown
(self)
slime/slime/utils/reloadable_process_group.py:169
Functionside_effect
(*args, **kwargs)
swe-rl/mini-swe-agent/tests/models/test_litellm_model.py:22
Functionsignal_handler
(sig, frame)
gui-rl/desktop_env/providers/aws/manager.py:67
Functionsignal_handler
(sig, frame)
gui-rl/desktop_env/providers/volcengine/manager.py:179
Functionsignal_handler
(sig, frame)
gui-rl/desktop_env/providers/aliyun/manager.py:64
Functionsigterm_handler
(_signo, _stack_frame)
Megatron-LM/tests/test_utils/python_scripts/launch_jet_workload.py:35
Methodsilent_error
(self, *args, **kwargs)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_local.py:327
Methodsimulate_crash
(self)
slime/slime/backends/sglang_utils/sglang_engine.py:479
Functionskip_first_line_load_json
(fp)
gui-rl/desktop_env/evaluators/metrics/vscode.py:27
Methodslice_teacher_payload
(value)
slime/slime/backends/megatron_utils/actor.py:1195
Functionsort_key
(x)
slime/slime/ray/placement_group.py:20
Methodspec_accept_length
(self)
slime/slime/utils/types.py:61
Methodspec_accept_rate
(self)
slime/slime/utils/types.py:57
Functionsplit
Split pipeline parallel size = 1 checkpoint to pipeline parallel size N.
Megatron-LM/examples/multimodal/nvlm/pp_checkpoint_converter.py:14
Methodsplit_sentences
(self, file_name)
Megatron-LM/tools/preprocess_data.py:131
Methodspread
(self)
slime/slime/utils/seqlen_balancing.py:70
Functionsquared_relu
(x)
Megatron-LM/megatron/training/yaml_arguments.py:420
Methodsquared_relu
(x)
Megatron-LM/megatron/legacy/model/transformer.py:142
Methodstart
(self)
openclaw-rl/openclaw_api_server.py:698
Methodstart
(self)
openclaw-rl/oel/openclaw_oel_api_server.py:1630
Methodstart
(self)
openclaw-opd/openclaw_opd_api_server.py:973
Methodstart
(self)
openclaw-tinker/api_server.py:437
Methodstart
Start the timer. Args: barrier (bool, optional): Synchronizes ranks before starting. Defaults to False.
Megatron-LM/megatron/core/timers.py:42
Methodstart
(self, barrier=False)
Megatron-LM/megatron/core/timers.py:84
Methodstart
(self)
slime/slime/utils/profile_utils.py:104
Methodstart
(self)
slime/slime/utils/profile_utils.py:135
← previousnext →8,901–9,000 of 10,561, ranked by callers