MCPcopy Create free account

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

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

↓ 1 callersFunctiontest_all_pipelines
Run all pipelines.
Megatron-LM/tests/functional_tests/test_cases/common/ckpt_converter/__main__.py:879
↓ 1 callersFunctiontest_allmasked_softmax_backward
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:346
↓ 1 callersFunctiontest_allmasked_softmax_forward
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:330
↓ 1 callersMethodtest_backward_consistency
Test that Triton and Python implementations produce same gradients.
slime/tests/test_fused_experts_backward.py:345
↓ 1 callersMethodtest_backward_consistency
Test that Triton and Python implementations produce same gradients.
slime/tests/test_fused_experts_backward.py:445
↓ 1 callersFunctiontest_basic_comparison
Test basic griffe comparison
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:111
↓ 1 callersFunctiontest_bin_reader
()
Megatron-LM/tests/unit_tests/data/test_bin_reader.py:147
↓ 1 callersFunctiontest_broadcast_data
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_data.py:14
↓ 1 callersFunctiontest_builder
()
Megatron-LM/tests/unit_tests/data/test_builder.py:51
↓ 1 callersMethodtest_capture_freeze_gc
Test cuda graph capture while freezing the GC.
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:673
↓ 1 callersFunctiontest_checker_script
Test that the checker script exists
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:56
↓ 1 callersFunctiontest_column_parallel_linear
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:173
↓ 1 callersMethodtest_constructor
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_grouped_mlp.py:286
↓ 1 callersMethodtest_constructor
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_sequential_mlp.py:51
↓ 1 callersFunctiontest_cross_entropy
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_cross_entropy.py:45
↓ 1 callersFunctiontest_cuda_rng_tracker
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_random.py:73
↓ 1 callersFunctiontest_decorator_module
Test that the compat decorator module exists
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:41
↓ 1 callersFunctiontest_decorators_work
Test that decorators can be applied
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:84
↓ 1 callersFunctiontest_domain_mcore
()
Megatron-LM/tests/unit_tests/test_imports.py:125
↓ 1 callersFunctiontest_fim_gpt_dataset
(spm_rate, split_sample)
Megatron-LM/tests/unit_tests/data/test_fim_dataset.py:15
↓ 1 callersMethodtest_forward_consistency
Test that Triton and Python implementations produce same forward output.
slime/tests/test_fused_experts_backward.py:321
↓ 1 callersMethodtest_forward_consistency
Test that Triton and Python implementations produce same forward output.
slime/tests/test_fused_experts_backward.py:413
↓ 1 callersFunctiontest_fused_softmax
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:25
↓ 1 callersFunctiontest_fused_upper_triangle_mask_softmax
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:123
↓ 1 callersFunctiontest_get_tensor_model_parallel_src_rank
(tensor_model_parallel_size_)
Megatron-LM/megatron/legacy/mpu/tests/test_initialize.py:49
↓ 1 callersMethodtest_gpu_cudagraph
(self)
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:522
↓ 1 callersMethodtest_gpu_forward
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_grouped_mlp.py:164
↓ 1 callersMethodtest_gpu_forward
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_sequential_mlp.py:59
↓ 1 callersMethodtest_gpu_forward_with_no_tokens_allocated
Test the case when no token is allocated for groupedGEMM kernels.
Megatron-LM/tests/unit_tests/transformer/moe/test_grouped_mlp.py:187
↓ 1 callersMethodtest_gpu_forward_with_no_tokens_allocated
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_sequential_mlp.py:200
↓ 1 callersMethodtest_gpu_forward_with_one_local_expert
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_sequential_mlp.py:169
↓ 1 callersFunctiontest_griffe_installed
Test that griffe is installed
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:26
↓ 1 callersFunctiontest_initialize_affine_weight
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:94
↓ 1 callersFunctiontest_initialize_model_parallel
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_initialize.py:11
↓ 1 callersFunctiontest_layer_norm
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:223
↓ 1 callersMethodtest_llava_cudagraph_is_last_layer_logic
Test that is_last_layer logic correctly resets prev_bwd_hidden_state_inputgrad for LLaVA models.
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:382
↓ 1 callersFunctiontest_load_fused_kernels
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:13
↓ 1 callersFunctiontest_loss_mask_qwen3_simple
(model_name: str = "Qwen/Qwen3-8B")
slime/tests/utils/test_mask_utils.py:6
↓ 1 callersFunctiontest_loss_mask_qwen3_tools
(model_name: str = "Qwen/Qwen3-8B")
slime/tests/utils/test_mask_utils.py:26
↓ 1 callersFunctiontest_masked_softmax_backward
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:309
↓ 1 callersFunctiontest_masked_softmax_forward
()
Megatron-LM/megatron/legacy/fused_kernels/tests/test_fused_kernels.py:294
↓ 1 callersFunctiontest_mock_gpt_dataset
()
Megatron-LM/tests/unit_tests/data/test_gpt_dataset.py:31
↓ 1 callersFunctiontest_mock_multimodal_dataset
()
Megatron-LM/tests/unit_tests/data/test_multimodal_dataset.py:20
↓ 1 callersFunctiontest_model_parallel_cuda_manual_seed
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_random.py:144
↓ 1 callersMethodtest_moe_partial_cudagraph
(self, ep_size, moe_dropless_dispatcher, moe_dispatcher_type)
Megatron-LM/tests/unit_tests/transformer/test_cuda_graphs.py:1238
↓ 1 callersFunctiontest_optimizer_states_match
Compare optimizer state across provided torch_dist checkpoints: - Keys: ensure the set of optimizer tensor keys match across checkpoints
Megatron-LM/tests/functional_tests/python_test_utils/test_optimizer_grads_match.py:291
↓ 1 callersFunctiontest_parallel_embedding
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:16
↓ 1 callersFunctiontest_parallel_self_attention
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:348
↓ 1 callersFunctiontest_parallel_transformer_layer
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:436
↓ 1 callersMethodtest_pause
Pause/resume test.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:299
↓ 1 callersMethodtest_pp
Simple test with no TP, but PP.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:267
↓ 1 callersFunctiontest_preprocess_data_bert
()
Megatron-LM/tests/unit_tests/data/test_preprocess_data.py:218
↓ 1 callersFunctiontest_preprocess_data_gpt
()
Megatron-LM/tests/unit_tests/data/test_preprocess_data.py:186
↓ 1 callersFunctiontest_preprocess_mmdata
()
Megatron-LM/tests/unit_tests/data/test_preprocess_mmdata.py:195
↓ 1 callersMethodtest_proxy
Test if proxy is working
gui-rl/desktop_env/providers/aws/proxy_pool.py:112
↓ 1 callersFunctiontest_row_parallel_linear
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_layers.py:240
↓ 1 callersFunctiontest_set_cuda_rng_state
(tensor_model_parallel_size)
Megatron-LM/megatron/legacy/mpu/tests/test_random.py:11
↓ 1 callersMethodtest_simple
Simple test with no TP or PP.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:251
↓ 1 callersMethodtest_throughput
Throughput test with no TP or PP.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:390
↓ 1 callersMethodtest_tp
Simple test with TP, but no PP.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:259
↓ 1 callersMethodtest_tp_pp
Simple test with both TP and PP.
Megatron-LM/tests/unit_tests/inference/test_data_parallel_inference_coordinator.py:275
↓ 1 callersMethodtest_weight_init_value_the_same
(self)
Megatron-LM/tests/unit_tests/transformer/moe/test_grouped_mlp.py:136
↓ 1 callersFunctiontest_workflow
Test that the GitHub Actions workflow exists
Megatron-LM/tests/unit_tests/test_api_backwards_compat_setup.py:70
↓ 1 callersMethodtext_to_bert
Text to Bert tokens.
Megatron-LM/tools/retro/cli/cli.py:87
↓ 1 callersMethodtext_to_ids
Convert a text to a list of IDs.
Megatron-LM/megatron/core/tokenizers/text/libraries/bytelevel_tokenizer.py:73
↓ 1 callersMethodtie_embeddings_and_output_weights_state_dict
Ties the embedding and output weights in a given sharded state dict. Args: sharded_state_dict (ShardedStateDict): state dict with
Megatron-LM/megatron/core/models/common/language_module/language_module.py:381
↓ 1 callersFunctiontie_word_embeddings_state_dict
tie the embedding of the mtp processing stage in a given sharded state dict. Args: sharded_state_dict (ShardedStateDict): state dict with
Megatron-LM/megatron/core/transformer/multi_token_prediction.py:60
↓ 1 callersMethodtimed_barrier
(timeout_s=10)
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:1014
↓ 1 callersFunctionto_cuda
(d)
Megatron-LM/tests/unit_tests/test_optimizer.py:82
↓ 1 callersFunctionto_empty_if_meta
Move tensors to device if not meta device; otherwise materialize with empty_like(). Args: module: The target module to apply this tran
Megatron-LM/megatron/post_training/utils.py:41
↓ 1 callersFunctionto_empty_if_meta_device
Move tensors to device if not meta device; otherwise materialize with empty_like(). Officially, torch suggests to_empty() for meta device materia
Megatron-LM/megatron/training/utils.py:720
↓ 1 callersMethodto_payload
(self)
terminal-rl/custom_types.py:19
↓ 1 callersMethodto_payload
(self)
terminal-rl/custom_types.py:47
↓ 1 callersMethodtoken_to_id
Convert a token to its corresponding ID.
Megatron-LM/megatron/core/tokenizers/text/libraries/bytelevel_tokenizer.py:109
↓ 1 callersMethodtoken_to_id
Converts a single token to it's id.
Megatron-LM/megatron/core/tokenizers/text/libraries/huggingface_tokenizer.py:222
↓ 1 callersMethodtoken_to_id
Converts a single token to it's id.
Megatron-LM/megatron/core/tokenizers/text/libraries/sentencepiece_tokenizer.py:246
↓ 1 callersMethodtokenize
(self, *text)
Megatron-LM/tools/preprocess_mmdata.py:41
↓ 1 callersMethodtokenize
Text tokenization. Args: text (str): text to be tokenized. Returns: list: list of ids.
Megatron-LM/megatron/core/tokenizers/text/text_tokenizer.py:64
↓ 1 callersMethodtokenize
Tokenize a string.
Megatron-LM/megatron/training/tokenizer/gpt2_tokenization.py:239
↓ 1 callersMethodtokenize
Tokenize conversation or string input.
Megatron-LM/megatron/training/tokenizer/multimodal_tokenizer.py:207
↓ 1 callersMethodtokenize_encoder_prompt
Utility to tokenize the encoder_prompt Args: encoder_prompt (str): The encoder_prompt tokenizer (_type_): Tokenizer u
Megatron-LM/megatron/core/inference/model_inference_wrappers/t5/t5_inference_wrapper.py:98
↓ 1 callersFunctiontokenize_prompts
Tokenize prompts and make them avaiable on all ranks.
Megatron-LM/tools/retro/text_generation/retro_api.py:18
↓ 1 callersFunctiontokenize_prompts
Tokenize prompts and make them avaiable on all ranks. Args: data_parallel (bool): Broadcast tokens across a single data parallel model re
Megatron-LM/megatron/core/inference/text_generation_server/tokenization.py:12
↓ 1 callersMethodtokens_to_text
Converts list of tokens text.
Megatron-LM/megatron/core/tokenizers/text/libraries/huggingface_tokenizer.py:217
↓ 1 callersFunctiontopk_accuracy
(k)
Megatron-LM/pretrain_ict.py:97
↓ 1 callersFunctiontopk_routing_with_score_function
Compute the routing probabilities and map for top-k selection with score function. Args: logits (torch.Tensor): Logits tensor. top
Megatron-LM/megatron/core/transformer/moe/moe_utils.py:530
↓ 1 callersFunctiontorch_chunk_gated_delta_rule
Torch-native implementation of chunked gated delta rule for deterministic mode. Need this because FLA is not deterministic. Reference: h
Megatron-LM/megatron/core/ssm/gated_delta_net.py:577
↓ 1 callersFunctiontorch_cross_entropy
(batch_size, seq_length, vocab_size, logits_scale, seed)
Megatron-LM/megatron/legacy/mpu/tests/test_cross_entropy.py:16
↓ 1 callersMethodtorch_linear_cross_entropy_single_gpu
( hidden: torch.Tensor, weight: torch.Tensor, labels: torch.Tensor, reduction:
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:582
↓ 1 callersMethodtorch_linear_cross_entropy_single_gpu
( hidden: torch.Tensor, weight: torch.Tensor, labels: torch.Tensor, reduction:
Megatron-LM/tests/unit_tests/fusions/test_fused_linear_cross_entropy.py:1034
↓ 1 callersMethodtrack_and_report_progress
Utility function for tracking progress
Megatron-LM/megatron/legacy/indexer.py:68
↓ 1 callersMethodtrack_indexer_metrics
Track the sparse attention indexer metrics for logging. Args: loss_scale: Scale factor for the loss. iteration: Curre
Megatron-LM/megatron/core/transformer/experimental_attention_variant/dsa.py:116
↓ 1 callersFunctiontrack_moe_metrics
Track the MoE metrics for logging.
Megatron-LM/megatron/core/transformer/moe/moe_utils.py:800
↓ 1 callersFunctiontrain
(args)
openclaw-rl/oel/train_async.py:74
↓ 1 callersFunctiontrain
Training function: run train_step desired number of times, run validation, checkpoint.
Megatron-LM/megatron/training/training.py:2105
↓ 1 callersFunctiontrain
(args)
slime/train_async.py:74
↓ 1 callersFunctiontrain
(args)
slime/train.py:9
↓ 1 callersMethodtrain
Train index. Args: config (RetroPreprocessingConfig): Retro preprocessing config.
Megatron-LM/megatron/core/datasets/retro/index/indexes/faiss_base.py:95
↓ 1 callersMethodtrain
Run one training update over a rollout batch. Parameters: rollout_id: Monotonic id for logging. rollout_data_ref: A B
slime/slime/backends/fsdp_utils/actor.py:467
↓ 1 callersMethodtrain_actor
(self, rollout_id: int, rollout_data: RolloutBatch)
slime/slime/backends/megatron_utils/actor.py:1170
← previousnext →5,201–5,300 of 10,561, ranked by callers