MCPcopy Create free account

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

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

Methodtest_forward
(self, ngroups_local_tp, cp_size, D_has_hdim)
Megatron-LM/tests/unit_tests/ssm/test_mamba_context_parallel.py:26
Methodtest_forward_backward
Test MTP forward and backward with gptmodel.
Megatron-LM/tests/unit_tests/transformer/test_multi_token_prediction.py:337
Methodtest_forward_backward
(self, tp_size, ep_size, permute_fusion, deterministic, monkeypatch)
Megatron-LM/tests/unit_tests/transformer/moe/test_a2a_token_dispatcher.py:34
Methodtest_forward_backward
(self, tp_size, ep_size, permute_fusion)
Megatron-LM/tests/unit_tests/transformer/moe/test_token_dispatcher.py:370
Methodtest_forward_backward
(self, tp_size, ep_size, permute_fusion, moe_flex_dispatcher_backend)
Megatron-LM/tests/unit_tests/transformer/moe/test_token_dispatcher.py:436
Methodtest_forward_backward_for_kv
(self, input_format)
Megatron-LM/tests/unit_tests/fusions/test_mla_yarn_rope_apply.py:254
Methodtest_forward_backward_for_q
(self, input_format)
Megatron-LM/tests/unit_tests/fusions/test_mla_yarn_rope_apply.py:251
Functiontest_forward_backward_func_with_full_cuda_graph
(mocker)
Megatron-LM/tests/unit_tests/transformer/test_full_cuda_graph.py:25
Functiontest_forward_backward_func_with_interleaving
(mocker)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:248
Functiontest_forward_backward_func_with_pipeline_parallel
(mocker)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:186
Functiontest_forward_backward_func_with_uneven_interleaving
(mocker)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:352
Functiontest_forward_backward_func_without_pipeline_parallel
(mocker)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:133
Functiontest_forward_backward_no_pipelining_with_custom_pgs
Validate no-pipeline schedule when explicit custom PGs are provided.
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:655
Functiontest_forward_backward_pipelining_with_interleaving_with_custom_pgs
Test that forward_backward_pipelining_with_interleaving produces the same output with and without explicit process group parameters.
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:561
Functiontest_forward_backward_pipelining_without_interleaving_with_custom_pgs
Test that forward_backward_pipelining_without_interleaving produces the same output with and without explicit process group parameters.
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:468
Methodtest_forward_fsdp
Test FSDP workaround for text-only data. FSDP can hang with text-only data. As a workaround, we run the vision model with a dummy image,
Megatron-LM/tests/unit_tests/models/test_llava_model.py:402
Methodtest_forward_output_encoder_hidden_only
(self)
Megatron-LM/tests/unit_tests/models/test_t5_model.py:111
Methodtest_forward_pass
Test the complete forward pass.
Megatron-LM/tests/unit_tests/models/test_mimo_submodules.py:282
Methodtest_forward_pass
Test that forward pass preserves output.
Megatron-LM/tests/unit_tests/transformer/test_attention_variant_dsa.py:218
Methodtest_forward_text_only
Test forward pass with only text input.
Megatron-LM/tests/unit_tests/models/test_mimo_model.py:261
Functiontest_forward_vpp
(create_args, tmp_path_dist_ckpt, tp_pp_vpp, pp_layout, is_moe, with_mtp)
Megatron-LM/tests/unit_tests/pipeline_parallel/test_pipeline_layout.py:210
Methodtest_forward_with_encoder_hidden_states
(self)
Megatron-LM/tests/unit_tests/models/test_t5_model.py:114
Methodtest_forward_with_image_and_audio_modality
Test forward pass with text, image, and audio input.
Megatron-LM/tests/unit_tests/models/test_mimo_model.py:360
Methodtest_forward_with_image_modality
Test forward pass with text and image input.
Megatron-LM/tests/unit_tests/models/test_mimo_model.py:291
Methodtest_fp16_module
(self)
Megatron-LM/tests/unit_tests/transformer/test_module.py:69
Methodtest_fp32_optimizer_resharding
(self, tmp_path_dist_ckpt, src_tp_pp, dest_tp_pp)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_optimizer.py:887
Methodtest_fp8_inference
(self, model_provider: str)
Megatron-LM/tests/unit_tests/inference/engines/test_dynamic_engine.py:804
Methodtest_fp8_save_load
( self, tmp_path_dist_ckpt, use_fpsl, src_tp_pp, dest_tp_pp, load_exchange_algo )
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fp8.py:76
Methodtest_fp8_support
Test MTP with FP8 training enabled.
Megatron-LM/tests/unit_tests/transformer/test_multi_token_prediction.py:439
Methodtest_freeze
(self)
Megatron-LM/tests/unit_tests/models/test_llava_model.py:452
Methodtest_from_rank_offsets_constructor
(self, dtype=torch.float, device='cuda')
Megatron-LM/tests/unit_tests/dist_checkpointing/test_mapping.py:29
Methodtest_from_rank_offsets_flat_constructor
(self, dtype=torch.float, device='cuda')
Megatron-LM/tests/unit_tests/dist_checkpointing/test_mapping.py:42
Functiontest_fsdp2_constructor
Test the FSDP2 constructor.
Megatron-LM/tests/unit_tests/distributed/test_torch_fully_sharded_parallel.py:49
Functiontest_fsdp2_constructor_with_process_group
Test the FSDP2 constructor with explicit process group parameter.
Megatron-LM/tests/unit_tests/distributed/test_torch_fully_sharded_parallel.py:77
Functiontest_fsdp_import
()
slime/tests/test_fsdp_import.py:4
Methodtest_fsdp_user_buffer_registration
Test that FSDP works correctly with user buffer registration. This test compares the training results of the baseline fsdp with the target fsd
Megatron-LM/tests/unit_tests/distributed/test_mcore_fully_sharded_data_parallel.py:225
Methodtest_fsdp_with_hybrid_sharding
Test that FSDP works correctly with hybrid sharding.
Megatron-LM/tests/unit_tests/distributed/test_mcore_fully_sharded_data_parallel.py:466
Methodtest_fsdp_with_process_groups
Test that FSDP works correctly with different process group configurations.
Megatron-LM/tests/unit_tests/distributed/test_mcore_fully_sharded_data_parallel.py:88
Methodtest_full_dp_reads
DP is the whole world.
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fully_parallel.py:475
Methodtest_fully_shard
Test the fully_shard API with different configurations. Does NOT test for performance or convergence. NOTE(@cspades): This t
Megatron-LM/tests/unit_tests/distributed/fsdp/test_mfsdp_fully_shard.py:228
Functiontest_func1
()
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:92
Functiontest_func2
()
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:99
Methodtest_fused_kernel_availability
Test is_kernel_available method with different configurations.
Megatron-LM/tests/unit_tests/fusions/test_torch_softmax.py:164
Methodtest_fused_rope_gpu_forward
(self)
Megatron-LM/tests/unit_tests/transformer/test_attention_packed_seq.py:105
Methodtest_fused_rope_gpu_forward
(self, rotary_interleaved, fused_qkv_rope)
Megatron-LM/tests/unit_tests/transformer/test_attention.py:145
Methodtest_fwd_bwd_pass_mix_and_match_transformer_blocks
(self)
Megatron-LM/tests/unit_tests/transformer/test_transformer_block_custom_pgs.py:475
Methodtest_fwd_bwd_pass_non_uniform_transformer_block
Test that parameters and gradients after one forward and backward pass with different process groups for attention and mlp.
Megatron-LM/tests/unit_tests/transformer/test_transformer_block_custom_pgs.py:382
Functiontest_gae_parallel_matches_serial
Test that chunked_gae (parallel-scan) matches vanilla_gae (batch-serial) under various shapes, chunk sizes and dtypes.
slime/tests/test_chunked_gae.py:17
Functiontest_gather_split_1d_tensor
()
Megatron-LM/tests/unit_tests/tensor_parallel/test_tensor_parallel_utils.py:31
Methodtest_gen_rank_enum_multiple_dims
Test _gen_rank_enum for multiple dimensions.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:102
Methodtest_gen_rank_enum_single_dim
Test _gen_rank_enum for single dimension.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:92
Methodtest_gen_rank_enum_with_offset
Test _gen_rank_enum with rank offset.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:112
Methodtest_generate_all_output_tokens_static_batch
(self)
Megatron-LM/tests/unit_tests/inference/text_generation_controllers/test_vlm_text_generation_controller.py:118
Methodtest_generate_all_output_tokens_static_batch
(self, dtype, symmetric_ar_type, fp8)
Megatron-LM/tests/unit_tests/inference/text_generation_controllers/test_simple_text_generation_controller.py:322
Methodtest_generate_all_output_tokens_static_batch
(self)
Megatron-LM/tests/unit_tests/inference/text_generation_controllers/test_encoder_decoder_text_generation_controller.py:111
Methodtest_generate_dynamic
(self, batch_size: int, num_trials: int, empty_prompt: bool)
Megatron-LM/tests/unit_tests/inference/engines/test_static_engine.py:170
Methodtest_generate_function
Test the generate function that processes multiple prompts at once.
Megatron-LM/tests/unit_tests/inference/engines/test_dynamic_engine.py:709
Methodtest_generate_legacy_static
(self, batch_size: int, num_trials: int, empty_prompt: bool)
Megatron-LM/tests/unit_tests/inference/engines/test_static_engine.py:135
Functiontest_generations_endpoint
(mock_send_do_generate, client, gpt2_tiktoken_tokenizer)
Megatron-LM/tests/unit_tests/test_inference.py:61
Methodtest_get
(self)
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:106
Methodtest_get
(self)
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:130
Methodtest_get_boundary_pp_stage_ranks
Test get_boundary_pp_stage_ranks function with different parallelism configurations.
Megatron-LM/tests/unit_tests/pipeline_parallel/test_bridge_communicator.py:420
Methodtest_get_cuda_graph_input_data
Test _get_cuda_graph_input_data function in TECudaGraphHelper.
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:765
Functiontest_get_current_global_batch_size
()
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:53
Methodtest_get_current_global_batch_size
(self)
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:109
Methodtest_get_current_global_batch_size
(self)
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:133
Methodtest_get_deterministic_model
Test that get_model can instantiate DeterministicModel via model_class parameter.
swe-rl/mini-swe-agent/tests/models/test_init.py:140
Methodtest_get_environment_class_invalid_module
Test that get_environment_class raises ValueError for non-existent module.
swe-rl/mini-swe-agent/tests/environments/test_init.py:23
Methodtest_get_environment_class_invalid_spec
Test that get_environment_class raises ValueError for invalid spec.
swe-rl/mini-swe-agent/tests/environments/test_init.py:18
Methodtest_get_environment_class_local_full_path
Test that get_environment_class returns LocalEnvironment when given full module path.
swe-rl/mini-swe-agent/tests/environments/test_init.py:8
Methodtest_get_environment_class_local_shorthand
Test that get_environment_class returns LocalEnvironment when given shorthand.
swe-rl/mini-swe-agent/tests/environments/test_init.py:13
Functiontest_get_forward_backward_func
()
Megatron-LM/tests/unit_tests/pipeline_parallel/test_schedules.py:37
Functiontest_get_gpt_modelopt_spec_interface
()
Megatron-LM/tests/unit_tests/post_training/test_modelopt_module_spec.py:227
Methodtest_get_grad_norm
Test LayerWiseDistributedOptimizer gradient norm computation.
Megatron-LM/tests/unit_tests/test_layer_wise_optimizer.py:185
Functiontest_get_image_name_with_complex_instance_id
Test get_image_name with complex instance_id containing multiple double underscores
swe-rl/mini-swe-agent/tests/run/test_swebench.py:84
Functiontest_get_image_name_with_existing_image_name
Test get_image_name when image_name is already provided
swe-rl/mini-swe-agent/tests/run/test_swebench.py:64
Functiontest_get_image_name_with_none_image_name
Test get_image_name when image_name is explicitly None
swe-rl/mini-swe-agent/tests/run/test_swebench.py:77
Functiontest_get_image_name_without_image_name
Test get_image_name when image_name needs to be constructed
swe-rl/mini-swe-agent/tests/run/test_swebench.py:70
Functiontest_get_key_per_thread_returns_same_key
()
swe-rl/mini-swe-agent/tests/models/test_anthropic_model.py:21
Methodtest_get_kvcache_utilization_stats_with_requests
Test get_kvcache_utilization_stats() with empty context and then with active requests.
Megatron-LM/tests/unit_tests/inference/test_wandb_logging.py:70
Methodtest_get_layer_name_without_prefix
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_layers.py:105
Methodtest_get_layer_offset
(self)
Megatron-LM/tests/unit_tests/transformer/test_transformer_layer.py:133
Methodtest_get_layer_offset_parametrized
Parametrized test for get_transformer_layer_offset with different configurations. Tests various combinations of first/last stage sett
Megatron-LM/tests/unit_tests/transformer/test_transformer_layer.py:226
Methodtest_get_leader_rank
Test get_leader_rank function with different parallelism configurations.
Megatron-LM/tests/unit_tests/pipeline_parallel/test_bridge_communicator.py:454
Functiontest_get_logprobs
Test that getting logprobs at least does not crash.
Megatron-LM/tests/unit_tests/test_rl_utils.py:60
Functiontest_get_logprobs_with_sequence_packing
Test that getting logprobs at least does not crash.
Megatron-LM/tests/unit_tests/test_rl_utils.py:74
Functiontest_get_lr_cosine
(mock_optimizer)
Megatron-LM/tests/unit_tests/test_optimizer_param_scheduler.py:139
Functiontest_get_lr_linear
(mock_optimizer)
Megatron-LM/tests/unit_tests/test_optimizer_param_scheduler.py:106
Functiontest_get_mamba_stack_modelopt_spec_interface
()
Megatron-LM/tests/unit_tests/post_training/test_modelopt_module_spec.py:266
Methodtest_get_megatron_muon_optimizer_layer_wise
Test get_megatron_muon_optimizer with layer-wise distributed optimizer.
Megatron-LM/tests/unit_tests/test_muon_optimizer.py:247
Methodtest_get_megatron_muon_optimizer_smoke
Smoke test for get_megatron_muon_optimizer function.
Megatron-LM/tests/unit_tests/test_muon_optimizer.py:132
Methodtest_get_megatron_muon_optimizer_validation
Test validation logic for get_megatron_muon_optimizer.
Megatron-LM/tests/unit_tests/test_muon_optimizer.py:207
Functiontest_get_megatron_optimizer_custom_process_groups_validation
Test validation logic for custom process groups in get_megatron_optimizer.
Megatron-LM/tests/unit_tests/test_optimizer.py:508
Functiontest_get_megatron_optimizer_with_custom_process_groups
Test that get_megatron_optimizer works correctly with custom process groups provided via pg_collection parameters.
Megatron-LM/tests/unit_tests/test_optimizer.py:380
Functiontest_get_micro_batch_size
()
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:62
Functiontest_get_mlp_module_spec_interface
()
Megatron-LM/tests/unit_tests/models/test_gpt_model.py:109
Methodtest_get_model_weights_converter
(self, mocker)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_single_device_converter.py:15
Methodtest_get_model_weights_converter
test model weights onverter
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_distributed_gpu_converter.py:57
Methodtest_get_model_weights_converter
(self, mocker)
Megatron-LM/tests/unit_tests/export/trtllm/test_single_device_fp8.py:217
Methodtest_get_model_weights_converter
(self, mocker)
Megatron-LM/tests/unit_tests/export/trtllm/test_distributed_fp8.py:220
← previousnext →9,601–9,700 of 10,561, ranked by callers