MCPcopy Create free account

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

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

Methodtest_transformer_layer_overlap
Verifies all-to-all overlap optimization in transformer layer produces the same results as the reference implementation.
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_layer_1f1b.py:405
Methodtest_transformer_layer_overlap_dense
Verifies all-to-all overlap optimization in dense transformer layer produces the same results as the reference implementation.
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_layer_1f1b.py:261
Methodtest_transformer_layer_overlap_early_attn_memory_release
Verifies all-to-all overlap optimization in transformer layer with early attn memory release produces the same results as the referen
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_layer_1f1b.py:351
Methodtest_transformer_layer_overlap_shared_expert
Verifies all-to-all overlap optimization in transformer layer with shared expert produces the same results as the reference implement
Megatron-LM/tests/unit_tests/a2a_overlap/test_schedule_layer_1f1b.py:298
Methodtest_transformer_process_groups
Test basic functionality of TransformerProcessGroups.
Megatron-LM/tests/unit_tests/test_process_groups_config.py:13
Methodtest_type_override
Test that argparse_meta can override argument type.
Megatron-LM/tests/unit_tests/test_argument_utils.py:473
Functiontest_uncaught_exception
(manager)
swe-rl/mini-swe-agent/tests/run/test_batch_progress.py:74
Methodtest_uneven_1d_sharding
(self, tmp_path_dist_ckpt, src_split, dest_split)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:688
Methodtest_uneven_1d_sharding_multiple_shards
The same as test_uneven_1d_sharding but with multiple shards per rank. src_split and dest_split have now 2 levels.
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:756
Methodtest_unexpected_keys_handling_during_validation
( self, caplog, tmp_path_dist_ckpt, validate_integrity, save_format )
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:854
Methodtest_unhandled_unsupported_callables
Test that an unsupported type produces a TypInferenceError.
Megatron-LM/tests/unit_tests/test_argument_utils.py:602
Methodtest_unhandled_unsupported_unions
Test that an unsupported type produces a TypInferenceError.
Megatron-LM/tests/unit_tests/test_argument_utils.py:622
Methodtest_unsafe_globals
(self, tmp_path_dist_ckpt)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_safe_globals.py:35
Methodtest_unset_api_key_scenario
Test unsetting an API key specifically.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:337
Methodtest_unset_command_via_typer
Test the unset command through the Typer app.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:451
Methodtest_unset_configured_flag
Test unsetting the configured flag.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:353
Methodtest_unset_existing_key
Test unsetting an existing key (legacy test for compatibility).
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:277
Methodtest_unset_from_empty_file
Test unsetting from an empty file.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:303
Methodtest_unset_from_file_with_multiple_keys
Test unsetting one key from a file with multiple keys.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:316
Methodtest_unset_nonexistent_key
Test unsetting a key that doesn't exist (should not error).
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:290
Methodtest_unset_with_argument_provided
Test unset command when key is provided as argument.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:245
Methodtest_unset_with_no_argument_prompts_for_key
Test unset command when no argument provided - should prompt for key.
swe-rl/mini-swe-agent/tests/run/test_extra_config.py:258
Methodtest_up_proj_recomputed_gpu_forward
(self)
Megatron-LM/tests/unit_tests/transformer/test_multi_latent_attention.py:345
Methodtest_upcycling_Local
(self, tp_ep, granularity, grouped_gemm, swiglu, squared_relu)
Megatron-LM/tests/unit_tests/transformer/moe/test_upcycling.py:187
Methodtest_upcycling_TE
(self, tp_ep, granularity, grouped_gemm, swiglu, squared_relu)
Megatron-LM/tests/unit_tests/transformer/moe/test_upcycling.py:257
Functiontest_update_instance_status
(manager)
swe-rl/mini-swe-agent/tests/run/test_batch_progress.py:82
Functiontest_update_num_microbatches
()
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:67
Functiontest_update_preds_file_existing_file
Test update_preds_file when output file already exists
swe-rl/mini-swe-agent/tests/run/test_swebench.py:191
Functiontest_update_preds_file_new_file
Test update_preds_file when output file doesn't exist
swe-rl/mini-swe-agent/tests/run/test_swebench.py:174
Functiontest_update_preds_file_overwrite_existing
Test update_preds_file overwrites existing instance
swe-rl/mini-swe-agent/tests/run/test_swebench.py:224
Methodtest_update_request
(self, is_hybrid_model: bool)
Megatron-LM/tests/unit_tests/inference/contexts/test_dynamic_context.py:573
Methodtest_validation_errors
Test validation errors when token counts don't match embedding counts.
Megatron-LM/tests/unit_tests/models/test_mimo_embedding_alignment.py:358
Functiontest_values_path
(request)
Megatron-LM/tests/functional_tests/python_test_utils/conftest.py:92
Methodtest_variable_batch_size_inference
(self)
Megatron-LM/tests/unit_tests/ssm/test_mamba_mixer.py:77
Functiontest_virtual_pipeline_model_parallel_rank
(order)
Megatron-LM/tests/unit_tests/test_parallel_state.py:160
Functiontest_vocab
()
Megatron-LM/tests/unit_tests/tensor_parallel/test_tensor_parallel_utils.py:43
Functiontest_vocab_parallel_cross_entropy
()
Megatron-LM/tests/unit_tests/tensor_parallel/test_cross_entropy.py:8
Methodtest_vocab_parallel_embedding_tp_attrs_no_init
(self, use_cpu_init)
Megatron-LM/tests/unit_tests/tensor_parallel/test_tp_attrs_without_init.py:20
Methodtest_vocab_size_padding_change
Test model loading with different vocab size (caused by TP padding).
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_bert_model.py:151
Methodtest_vocab_size_padding_change
Test model loading with different vocab size (caused by TP padding).
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_gpt_model.py:173
Functiontest_weighted_bias_swiglu
(input_dtype)
Megatron-LM/tests/unit_tests/fusions/test_swiglu_fusion.py:8
Functiontest_weighted_squared_relu_fusion
(input_dtype)
Megatron-LM/tests/unit_tests/fusions/test_weighted_squared_relu_fusion.py:13
Functiontest_whitelist_actions_bypass_confirmation
Test that whitelisted actions bypass confirmation.
swe-rl/mini-swe-agent/tests/agents/test_interactive_textual.py:381
Functiontest_whitelisted_actions_skip_confirmation
Test that whitelisted actions don't require confirmation.
swe-rl/mini-swe-agent/tests/agents/test_interactive.py:104
Methodtest_window_attention_integration
Test FusedScaleMaskSoftmax with window attention configuration.
Megatron-LM/tests/unit_tests/fusions/test_torch_softmax.py:148
Methodtest_window_size_different_formats
Test with different window_size formats.
Megatron-LM/tests/unit_tests/transformer/test_utils.py:284
Methodtest_window_size_with_no_skip_freq
Test when window_size is set but window_attn_skip_freq is None.
Megatron-LM/tests/unit_tests/transformer/test_utils.py:137
Methodtest_with_custom_process_groups
Test MambaModel with custom process groups.
Megatron-LM/tests/unit_tests/models/test_mamba_model.py:144
Methodtest_with_transformer_config_validation
Test that function works with fully validated TransformerConfig objects.
Megatron-LM/tests/unit_tests/transformer/test_utils.py:384
Functiontest_write_metadata
()
Megatron-LM/tests/unit_tests/tokenizers/test_tokenizer.py:221
Methodtest_wrong_length_override_pattern
(self)
Megatron-LM/tests/unit_tests/ssm/test_mamba_hybrid_layer_allocation.py:70
Methodtest_wrong_number_of_layer_types_in_override_pattern
(self)
Megatron-LM/tests/unit_tests/ssm/test_mamba_hybrid_layer_allocation.py:75
Functiontest_yaml_report_generation
(manager_with_yaml)
swe-rl/mini-swe-agent/tests/run/test_batch_progress.py:87
Functiontest_yolo_mode_activation
Test entering yolo mode disables confirmations.
swe-rl/mini-swe-agent/tests/agents/test_interactive.py:55
Functiontest_yolo_mode_confirms_pending_action
Test that pressing 'y' to switch to YOLO mode also confirms any pending action.
swe-rl/mini-swe-agent/tests/agents/test_interactive_textual.py:480
Functiontest_yolo_mode_sets_correct_agent_config
Test that yolo mode sets the correct agent configuration.
swe-rl/mini-swe-agent/tests/run/test_cli_integration.py:290
Methodtest_zero_parameters
(self)
Megatron-LM/tests/unit_tests/models/test_base_embedding.py:33
Methodtest_zero_tokens_generated_batch_vs_single
Verifies that when `num_tokens_to_generate=0`, the outputs from batched inference match the outputs from single-request inference for
Megatron-LM/tests/unit_tests/inference/text_generation_controllers/test_simple_text_generation_controller.py:633
Methodtext_to_ids
Converts text to tokens ids.
Megatron-LM/megatron/core/tokenizers/text/libraries/huggingface_tokenizer.py:236
Methodtext_to_ids
Converts text to tokens ids.
Megatron-LM/megatron/core/tokenizers/text/libraries/sentencepiece_tokenizer.py:141
Methodtext_to_ids
Converts text to ids. Args: text (str): text to be tokenized. Returns: List[int]: list of ids.
Megatron-LM/megatron/core/tokenizers/text/libraries/abstract_tokenizer.py:65
Methodtext_to_ids
Converts text to list of ids.
Megatron-LM/megatron/core/tokenizers/text/libraries/tiktoken_tokenizer.py:197
Methodtext_to_tokens
Convert a text to a list of tokens.
Megatron-LM/megatron/core/tokenizers/text/libraries/bytelevel_tokenizer.py:61
Methodtext_to_tokens
Converts text to tokens. Args: text (str): text to be tokenized. Returns: List[str]: list of tokens
Megatron-LM/megatron/core/tokenizers/text/libraries/abstract_tokenizer.py:13
Methodtext_to_tokens
Converts text to tokens.
Megatron-LM/megatron/core/tokenizers/text/libraries/tiktoken_tokenizer.py:160
Functiontie_output_layer_state_dict
tie the output layer of the mtp processing stage in a given sharded state dict. Args: sharded_state_dict (ShardedStateDict): state dict w
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:95
Functiontimed
Timing decorator.
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:35
Functiontimed_dec
(fn)
Megatron-LM/megatron/core/dist_checkpointing/strategies/two_stage.py:38
Functiontmp_dir_per_class
(tmp_path_factory)
Megatron-LM/tests/unit_tests/dist_checkpointing/conftest.py:15
Functiontmp_path_dist_ckpt
Common directory for saving the checkpoint. Can't use pytest `tmp_path_factory` directly because directory must be shared between processes.
Megatron-LM/tests/unit_tests/conftest.py:57
Functiontmp_path_dist_ckpt
Common directory for saving the checkpoint. Can't use pytest `tmp_path_factory` directly because directory must be shared between processes.
Megatron-LM/tests/unit_tests/transformer/moe/conftest.py:35
Methodto_dict
(self)
slime/slime/utils/types.py:70
Methodto_dict
(self)
slime/slime/utils/types.py:103
Methodtoken_combine
Combines expert outputs using Reduce-Scatter. This method performs the ReduceScatter communication operation to collect expert output
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:330
Methodtoken_combine
Executes fused un-permutation and communication using DeepEP kernels. This method performs the inverse AlltoAll communication operation to co
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:790
Methodtoken_combine
Executes fused un-permutation and communication using DeepEP kernels. This is the inverse of the `token_dispatch` operation. Args:
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:1490
Methodtoken_counter
(self)
terminal-rl/agent/camel_agent.py:262
Methodtoken_dispatch
Gathers tokens from all TP*EP ranks using AllGather.
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:258
Methodtoken_dispatch
Perform all-to-all communication for dispatching tokens. This method performs the all-to-all communication step to dispatch tokens a
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:650
Methodtoken_dispatch
Execute fused permutation and AlltoAll communication. This method currently leverages DeepEP's fused dispatch kernel, which combines
Megatron-LM/megatron/core/transformer/moe/token_dispatcher.py:1433
Methodtokenize
Convert text to embedding ids Args: text (str): The text to convert Returns: numpy.ndarray: The converted em
Megatron-LM/megatron/core/datasets/megatron_tokenizer.py:43
Methodtokenize
(self, text, **kwargs)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:163
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:248
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:353
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:463
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:551
Methodtokenize
Default args for text completion, not chat/dialog.
Megatron-LM/megatron/training/tokenizer/tokenizer.py:598
Methodtokenize
(self, s: str, bos: bool = False, eos: bool = False)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:776
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/tokenizer.py:891
Methodtokenize
Tokenize conversation or string input.
Megatron-LM/megatron/training/tokenizer/sft_tokenizer.py:139
Methodtokenize
(self, text)
Megatron-LM/megatron/training/tokenizer/bert_tokenization.py:170
Methodtokenize
Tokenizes a piece of text.
Megatron-LM/megatron/training/tokenizer/bert_tokenization.py:228
Methodtokenize
Tokenizes a piece of text into its word pieces. This uses a greedy longest-match-first algorithm to perform tokenization using the gi
Megatron-LM/megatron/training/tokenizer/bert_tokenization.py:340
Methodtokens_to_ids
Convert a list of tokens to a list of IDs.
Megatron-LM/megatron/core/tokenizers/text/libraries/bytelevel_tokenizer.py:87
Methodtokens_to_ids
Converts tokens to ids. Args: tokens (List[str]): tokens to be converted. Returns: List[int]: ids o
Megatron-LM/megatron/core/tokenizers/text/libraries/abstract_tokenizer.py:39
Methodtokens_to_ids
Converts list of tokens to list of it's ids.
Megatron-LM/megatron/core/tokenizers/text/libraries/tiktoken_tokenizer.py:177
Methodtokens_to_text
Convert a list of tokens to a text.
Megatron-LM/megatron/core/tokenizers/text/libraries/bytelevel_tokenizer.py:67
Methodtokens_to_text
Converts list of tokens text.
Megatron-LM/megatron/core/tokenizers/text/libraries/sentencepiece_tokenizer.py:219
Methodtokens_to_text
Converts tokens to text. Args: tokens (List[str]): tokens to be detokenized. Returns: str: detokeni
Megatron-LM/megatron/core/tokenizers/text/libraries/abstract_tokenizer.py:26
Methodtokens_to_text
Converts list of tokens to text.
Megatron-LM/megatron/core/tokenizers/text/libraries/tiktoken_tokenizer.py:165
← previousnext →10,301–10,400 of 10,561, ranked by callers