MCPcopy Create free account

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

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

Methodtest_optimizer_params
(self, tmp_path_dist_ckpt)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_optimizer.py:212
Functiontest_optimizer_reload_model_params
()
Megatron-LM/tests/unit_tests/test_optimizer.py:304
Methodtest_optimizer_resharding
( self, tmp_path_dist_ckpt, src_tp_pp, dest_tp_pp, use_dist_opt,
Megatron-LM/tests/unit_tests/dist_checkpointing/test_optimizer.py:971
Methodtest_optional_fields
Test that Optional fields are handled correctly.
Megatron-LM/tests/unit_tests/test_argument_utils.py:239
Methodtest_order_dims_all_dims
Test _order_dims with all dimensions.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:81
Methodtest_order_dims_multiple_dims
Test _order_dims with multiple dimensions.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:66
Methodtest_order_dims_single_dim
Test _order_dims with single dimension.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:55
Methodtest_out_of_order_load
DP group has 8 rank (TP=1).
Megatron-LM/tests/unit_tests/dist_checkpointing/test_fully_parallel.py:514
Methodtest_output_log_probs
(self, dtype)
Megatron-LM/tests/unit_tests/inference/text_generation_controllers/test_simple_text_generation_controller.py:400
Methodtest_output_shape
(self)
Megatron-LM/tests/unit_tests/fusions/test_torch_softmax.py:26
Methodtest_output_shape
(self)
Megatron-LM/tests/unit_tests/fusions/test_torch_softmax.py:62
Methodtest_output_shapes_match
(self)
Megatron-LM/tests/unit_tests/inference/test_flash_decode.py:15
Methodtest_packed_sequences
Test MTP with packed sequences.
Megatron-LM/tests/unit_tests/transformer/test_multi_token_prediction.py:475
Methodtest_padding_mechanism_works
Test that the padding mechanism actually pads and unpads correctly.
Megatron-LM/tests/unit_tests/test_fp8_utils.py:68
Functiontest_parallel_attention_correctness
( tmp_path_dist_ckpt, apply_rope_fusion, tp, sp, cp, qk_layernorm, fallback_to_eag
Megatron-LM/tests/unit_tests/transformer/test_attention.py:709
Functiontest_parallel_gated_delta_net_correctness
(tmp_path_dist_ckpt, tp, sp, cp)
Megatron-LM/tests/unit_tests/ssm/test_gated_delta_net.py:149
Methodtest_parallel_inference
( self, model_provider, tp_size, pp_size, ep_size, sequence_pa
Megatron-LM/tests/unit_tests/inference/engines/test_dynamic_engine.py:842
Methodtest_parallel_inference
(self, tp_size, pp_size, ep_size, sequence_parallel)
Megatron-LM/tests/unit_tests/inference/engines/test_static_engine.py:315
Functiontest_parallel_multi_latent_attention_correctness
( tmp_path_dist_ckpt, rope_type, apply_rope_fusion, tp, sp, cp )
Megatron-LM/tests/unit_tests/transformer/test_multi_latent_attention.py:1285
Methodtest_parallel_reconfiguration_e2e
Test model saving and loading with different TP/PP
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_t5_model.py:142
Methodtest_parallel_reconfiguration_e2e
Test module saving and loading with different TP/PP
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_mlp_glu.py:57
Methodtest_parallel_reconfiguration_e2e
Test model saving and loading with different TP/PP/EP/ETP(expert-tensor-parallel)
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_moe_experts.py:147
Methodtest_parallel_reconfiguration_e2e
Test model saving and loading with different TP/PP/expert parallelism
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_mamba.py:86
Methodtest_parallel_reconfiguration_e2e
Test model saving and loading with different TP/PP
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_bert_model.py:120
Methodtest_parallel_reconfiguration_e2e
Test model saving and loading with different TP/PP
Megatron-LM/tests/unit_tests/dist_checkpointing/models/test_gpt_model.py:131
Functiontest_param_norm
(use_distributed_optimizer: bool)
Megatron-LM/tests/unit_tests/test_utils.py:241
Methodtest_parameter_sharding
Test that parameters are correctly sharded across DP ranks.
Megatron-LM/tests/unit_tests/dist_checkpointing/test_layer_wise_optimizer.py:131
Methodtest_parameter_updates
Test LayerWiseDistributedOptimizer actually updates model parameters.
Megatron-LM/tests/unit_tests/test_layer_wise_optimizer.py:394
Methodtest_parameter_updates_insufficient_parameters
Test LayerWiseDistributedOptimizer when there are insufficient parameters for all ranks. Uses a tiny model with only 1 layer (2 parameters: w
Megatron-LM/tests/unit_tests/test_layer_wise_optimizer.py:398
Methodtest_params_and_grads_match_transformer_block
Test that parameters and gradients match after one forward and backward pass between transformer blocks using default and custom proc
Megatron-LM/tests/unit_tests/transformer/test_transformer_block_custom_pgs.py:227
Functiontest_parse_action_failures
Test action parsing raises appropriate exceptions for invalid formats.
swe-rl/mini-swe-agent/tests/agents/test_default.py:142
Functiontest_parse_action_success
Test action parsing works correctly for valid formats.
swe-rl/mini-swe-agent/tests/agents/test_default.py:121
Functiontest_parse_qlinear_params_example
()
Megatron-LM/tests/unit_tests/transformer/test_quantization_config.py:48
Methodtest_parsing_layout_from_str
(self, pp_size, input_layout_str, input_layout_list)
Megatron-LM/tests/unit_tests/transformer/test_transformer_block.py:608
Methodtest_partial_matches
Test that partial string matches work correctly.
swe-rl/mini-swe-agent/tests/models/test_init.py:63
Methodtest_partition_change_save_load
(self, tmp_path_dist_ckpt, strategy=None)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:197
Methodtest_partition_change_save_load
(self, tmp_path_dist_ckpt, src_tp_pp, dest_tp_pp)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_flattened_resharding.py:43
Methodtest_passthrough
(self, caplog, algo, validate_integrity, missing, unexpected, strict)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_strict.py:230
Methodtest_passthrough_errors
( self, caplog, algo, validate_integrity, missing, unexpected, strict )
Megatron-LM/tests/unit_tests/dist_checkpointing/test_strict.py:262
Methodtest_paused_requests_in_stats
Test that paused requests are correctly reflected in stats.
Megatron-LM/tests/unit_tests/inference/test_wandb_logging.py:228
Methodtest_performance
(self, problem, dtype, reduction, ignore_index)
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:386
Methodtest_performance
(self, problem, dtype, reduction)
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:803
Methodtest_performance
(self, problem, dtype, reduction)
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:1302
Methodtest_performance_with_many_layers
Test performance with a large number of layers.
Megatron-LM/tests/unit_tests/transformer/test_utils.py:403
Methodtest_pg_input_args
Test TransformerBlock with custom process groups.
Megatron-LM/tests/unit_tests/transformer/test_transformer_block.py:257
Functiontest_pipeline_model_parallel_rank
(order)
Megatron-LM/tests/unit_tests/test_parallel_state.py:121
Functiontest_pipeline_model_parallel_world_size
(order)
Megatron-LM/tests/unit_tests/test_parallel_state.py:94
Functiontest_pipeline_parallel_initializations
(order)
Megatron-LM/tests/unit_tests/test_parallel_state.py:55
Functiontest_placeholder
This is here because otherwise there's no other test in this module (all disabled) and pytest would fail.
Megatron-LM/tests/unit_tests/transformer/moe/test_a2a_token_dispatcher.py:15
Methodtest_post_process_forward
(self)
Megatron-LM/tests/unit_tests/models/test_t5_model.py:108
Methodtest_post_process_forward
(self)
Megatron-LM/tests/unit_tests/models/test_bert_model.py:76
Methodtest_post_process_forward
(self, heterogeneous_gpt_model, expected_num_parameters)
Megatron-LM/tests/unit_tests/models/test_heterogeneous_gpt_model.py:131
Methodtest_post_process_forward
(self)
Megatron-LM/tests/unit_tests/models/test_gpt_model.py:86
Methodtest_post_process_forward
(self)
Megatron-LM/tests/unit_tests/transformer/test_utils.py:53
Functiontest_precision_aware_fused_adam
()
Megatron-LM/tests/unit_tests/test_optimizer.py:97
Functiontest_precision_aware_optimizer
( precision: str, main_params_dtype: torch.dtype, main_grads_dtype: torch.dtype, moment_dtype:
Megatron-LM/tests/unit_tests/test_optimizer.py:155
Functiontest_prepare_data_for_update
Test that getting logprobs at least does not crash.
Megatron-LM/tests/unit_tests/test_rl_utils.py:309
Methodtest_prepare_model_for_fp8_inference_basic
Test prepare_model_for_fp8_inference wraps TE modules.
Megatron-LM/tests/unit_tests/test_fp8_utils.py:35
Functiontest_prepare_trajectories
()
Megatron-LM/tests/unit_tests/test_rl_utils.py:89
Functiontest_prepare_trajectories_with_packing
Test that rollouts data is properly prepared with sequence packing enabled.
Megatron-LM/tests/unit_tests/test_rl_utils.py:139
Functiontest_prepare_trajectories_with_sequence_packing
Test prepare_trajectories with sequence packing enabled.
Megatron-LM/tests/unit_tests/test_rl_utils.py:533
Methodtest_preprocess_data
(self)
Megatron-LM/tests/unit_tests/models/test_llava_model.py:90
Functiontest_print_report
Test that print_report produces expected output.
swe-rl/mini-swe-agent/tests/run/test_batch_progress.py:109
Methodtest_prints_both_limits_when_both_set
Test that both limits are printed when both environment variables are set.
swe-rl/mini-swe-agent/tests/models/test_init.py:166
Methodtest_prints_call_limit_when_set
Test that call limit is printed when MSWEA_GLOBAL_CALL_LIMIT is set.
swe-rl/mini-swe-agent/tests/models/test_init.py:159
Methodtest_prints_cost_limit_when_set
Test that cost limit is printed when MSWEA_GLOBAL_COST_LIMIT is set.
swe-rl/mini-swe-agent/tests/models/test_init.py:152
Methodtest_process_embedding_token_parallel
Test _process_embedding_token_parallel. Note: This test requires TE version >= 1.10.0 to run properly.
Megatron-LM/tests/unit_tests/models/test_llava_model.py:697
Methodtest_process_save_load
(self, tmp_path_dist_ckpt)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_msc.py:29
Functiontest_python_m_minisweagent_help
Test that python -m minisweagent --help works correctly.
swe-rl/mini-swe-agent/tests/run/test_cli_integration.py:422
Methodtest_raise_all
(self, validate_integrity, algo, missing, unexpected)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_strict.py:301
Methodtest_raise_unexpected
(self, validate_integrity, algo, missing)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_strict.py:283
Methodtest_raises_error_when_no_model_configured
Test that ValueError is raised when no model is configured anywhere.
swe-rl/mini-swe-agent/tests/models/test_init.py:34
Functiontest_ramp_up
()
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:137
Methodtest_rank_enumeration_correctness
Test that rank enumeration produces correct pattern.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:293
Functiontest_rank_generator_for_tp_dp_pp
(nodes, num_gpu, tp, pp, cp, ep)
Megatron-LM/tests/unit_tests/test_parallel_state.py:312
Functiontest_read_metadata_non_distributed
Test read_metadata without torch.distributed initialized.
Megatron-LM/tests/unit_tests/test_checkpointing.py:367
Methodtest_real_distributed_all_reduce
Test actual communication using the created process groups.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:420
Methodtest_real_distributed_basic_functionality
Test basic HyperCommGrid functionality with real distributed backend.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:330
Methodtest_real_distributed_different_world_sizes
Test HyperCommGrid with different valid world sizes.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:448
Methodtest_real_distributed_error_handling
Test error handling with real distributed backend.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:489
Methodtest_real_distributed_multi_dimensional_groups
Test multi-dimensional process group creation with real distributed backend.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:394
Methodtest_real_distributed_process_group_creation
Test process group creation with real distributed backend.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:359
Methodtest_real_distributed_rank_enumeration_verification
Verify rank enumeration produces correct communication patterns.
Megatron-LM/tests/unit_tests/test_hyper_comm_grid.py:516
Functiontest_recipe_config_matching
()
Megatron-LM/tests/unit_tests/transformer/test_quantization_config.py:22
Functiontest_reconfigure_num_microbatches_calculator
()
Megatron-LM/tests/unit_tests/test_num_microbatches_calculator.py:30
Functiontest_redo_existing_false_skips_existing
Test that redo_existing=False skips instances that already have results
swe-rl/mini-swe-agent/tests/run/test_swebench.py:298
Functiontest_redo_existing_true_overwrites_existing
Test that redo_existing=True processes instances even if they already have results
swe-rl/mini-swe-agent/tests/run/test_swebench.py:333
Methodtest_reduce_scatter_with_fp32_accumulation
( self, async_op: bool, baseline_reduce_scatter_in_fp32: bool )
Megatron-LM/tests/unit_tests/distributed/test_reduce_scatter_with_fp32_accumulation.py:33
Methodtest_reformulate_nd_flattened_tensors
( self, tmp_path_dist_ckpt, src_tp_pp, dest_tp_pp, expected_ckpt_offsets_by_rank )
Megatron-LM/tests/unit_tests/dist_checkpointing/test_flattened_resharding.py:86
Functiontest_regular_pipeline
( compare_approximate_results: bool, golden_values: Dict[str, common.GoldenValueMetric], actual_va
Megatron-LM/tests/functional_tests/python_test_utils/test_pretraining_regular_pipeline.py:25
Methodtest_release_memory_blocks_for_finished_requests
Test that memory blocks are correctly released for finished requests.
Megatron-LM/tests/unit_tests/inference/contexts/test_dynamic_context.py:772
Functiontest_remove_from_preds_file_existing
Test remove_from_preds_file removes existing instance
swe-rl/mini-swe-agent/tests/run/test_swebench.py:252
Functiontest_remove_from_preds_file_no_file
Test remove_from_preds_file when file doesn't exist
swe-rl/mini-swe-agent/tests/run/test_swebench.py:286
Functiontest_remove_from_preds_file_nonexistent_instance
Test remove_from_preds_file with nonexistent instance
swe-rl/mini-swe-agent/tests/run/test_swebench.py:271
Methodtest_remove_sharded_tensors
(self, tmp_path_dist_ckpt)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_serialization.py:552
Methodtest_rename_input_layer_names_to_trtllm_layer_names_exception
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_layers.py:31
Methodtest_rename_input_layer_names_to_trtllm_layer_names_with_layer_numbers
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_layers.py:76
Methodtest_rename_input_layer_names_to_trtllm_layer_names_without_layer_numbers
(self)
Megatron-LM/tests/unit_tests/export/trtllm/test_trtllm_layers.py:8
Functiontest_render_template_model_stats
Test that render_template has access to n_model_calls and model_cost from model.
swe-rl/mini-swe-agent/tests/agents/test_default.py:235
Methodtest_repl_save_and_load
(self, tmp_dir_per_class, tp, pp, async_save, algo, repl_groups)
Megatron-LM/tests/unit_tests/dist_checkpointing/test_replication.py:123
← previousnext →10,001–10,100 of 10,561, ranked by callers