MCPcopy Create free account

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

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

↓ 1 callersFunctionem_check
(prediction, golden_answers)
slime/examples/search-r1/qa_em_format.py:38
↓ 1 callersFunctionembed_block
Embed block of chunks. Args: config (RetroPreprocessingConfig): Retro preprocessing config. gpt_dataset (GPTChunkDataset): Chunk
Megatron-LM/megatron/core/datasets/retro/query/query.py:97
↓ 1 callersMethodembed_block
Embed a batch of tokens using the block model
Megatron-LM/megatron/legacy/model/realm_model.py:82
↓ 1 callersFunctionembed_data_loader
Iterate data loader and compute embeddings.
Megatron-LM/tools/bert_embedding/embed.py:96
↓ 1 callersMethodembed_query
Embed a batch of tokens using the query model
Megatron-LM/megatron/legacy/model/realm_model.py:73
↓ 1 callersMethodembed_text
Embed a single text string. Primarily used for on-the-fly embeddings, particularly during analysis or debugging. For large scale, use
Megatron-LM/tools/bert_embedding/embed.py:205
↓ 1 callersMethodembed_text_blocks
Process a text dataset in blocks.
Megatron-LM/tools/bert_embedding/embed.py:227
↓ 1 callersMethodembed_text_dataset
(self, text_dataset, verbose=True)
Megatron-LM/tools/bert_embedding/huggingface.py:81
↓ 1 callersFunctionembed_training_chunks
Embed DB chunks. Store chunks in blocks on disk. These blocks will later be merged into a single dataset for training the index. Args:
Megatron-LM/megatron/core/datasets/retro/index/build.py:160
↓ 1 callersMethodemit
(self, record: logging.LogRecord)
swe-rl/mini-swe-agent/src/minisweagent/agents/interactive_textual.py:107
↓ 1 callersFunctionempty_dir
(path: Path)
Megatron-LM/tests/unit_tests/dist_checkpointing/__init__.py:18
↓ 1 callersFunctionems
(prediction, ground_truths)
Megatron-LM/tools/retro/text_generation/evaluate.py:118
↓ 1 callersFunctionenable_debug_logging
Enable debug logging for PPTX comparison
gui-rl/desktop_env/evaluators/metrics/slides.py:15
↓ 1 callersMethodenable_decode_mode
Indicates the generation loop is in the decode phase (generating new output tokens). This should only be enabled if the generation lo
Megatron-LM/megatron/core/inference/contexts/static_context.py:62
↓ 1 callersFunctionenable_forward_pre_hook
(model_chunks)
Megatron-LM/tests/unit_tests/test_fp8_param.py:47
↓ 1 callersFunctionenable_jit_fuser
Enable the JIT fuser
Megatron-LM/megatron/core/jit.py:16
↓ 1 callersMethodenable_prefill_mode
Indicates the generation loop is in the prefill phase (still processing input prompt tokens). This should be enabled if the generatio
Megatron-LM/megatron/core/inference/contexts/static_context.py:54
↓ 1 callersMethodencode
Encode image data batch into a list of tensors. Args: encoders_data_batch: Dictionary containing encoder-specific inputs.
Megatron-LM/megatron/core/models/mimo/submodules/vision.py:55
↓ 1 callersMethodencode
Encode audio data into a sequence of embeddings. Args: encoders_data_batch: Dictionary containing encoder-specific inputs.
Megatron-LM/megatron/core/models/mimo/submodules/audio.py:47
↓ 1 callersMethodencode
(self, data_batch: Dict)
Megatron-LM/tests/unit_tests/models/test_mimo_submodules.py:31
↓ 1 callersMethodencode_batch_avlm_clip_whisper_llava
(self, batch_data: Dict)
Megatron-LM/examples/mimo/data/energon_avlm_task_encoder.py:313
↓ 1 callersMethodencode_batch_vlm_clip_llava
(self, batch_data: Dict)
Megatron-LM/examples/mimo/data/energon_vlm_task_encoder.py:220
↓ 1 callersMethodencode_batch_vlm_clip_llava_video
(self, batch_data: Dict)
Megatron-LM/examples/mimo/data/energon_vlm_task_encoder.py:245
↓ 1 callersMethodencode_captioning
Encode CaptioningSample.
Megatron-LM/examples/multimodal/dataset_helpers.py:247
↓ 1 callersMethodencode_lease
(worker_idx: int, worker_lease: str)
terminal-rl/router_server.py:74
↓ 1 callersMethodencode_llava_pretrain
Encode pretrain sample in LLAVA style.
Megatron-LM/examples/multimodal/dataset_helpers.py:293
↓ 1 callersMethodencode_ocr_prompt
Encode OCR sample.
Megatron-LM/examples/multimodal/dataset_helpers.py:772
↓ 1 callersMethodencode_ocr_ref_prompt
Encode OCR sample.
Megatron-LM/examples/multimodal/dataset_helpers.py:724
↓ 1 callersMethodencode_pdf_prompt
Encode OCR sample.
Megatron-LM/examples/multimodal/dataset_helpers.py:703
↓ 1 callersMethodencode_sample_list
We encode the list of samples using encode_llava_sft on each sample.
Megatron-LM/examples/multimodal/dataset_helpers.py:327
↓ 1 callersFunctionencoder_model_with_local_spec
T5 encoder local spec (uses Megatron-Core components).
Megatron-LM/megatron/core/models/T5/t5_spec.py:120
↓ 1 callersFunctionencoder_model_with_transformer_engine_default_spec
T5 encoder TE spec (uses Transformer Engine components).
Megatron-LM/megatron/core/models/T5/t5_spec.py:50
↓ 1 callersMethodend_recording
(self, lease_id: str, out_path: str)
gui-rl/env_client.py:89
↓ 1 callersMethodend_recording
(self, lease_id: str, out_path: str)
gui-rl/env_pool_server.py:258
↓ 1 callersFunctionensure_dataset_list
Normalize OmegaConf containers into a list of dicts. Accepts either a list or dictionary keyed by dataset name.
slime/slime/utils/eval_config.py:154
↓ 1 callersFunctionensure_divisibility
Ensure that numerator is divisible by the denominator.
Megatron-LM/megatron/core/utils.py:477
↓ 1 callersFunctionensure_homework_dir
If target_name dir doesn't exist in workspace, copy source_name to target_name.
openclaw-test/teacher_chat.py:122
↓ 1 callersFunctionensure_homework_dir
If target_name dir doesn't exist in workspace, copy source_name to target_name.
openclaw-test/TA_chat.py:117
↓ 1 callersMethodensure_required_arguments
Ensure that certain Megatron arguments (from checkpoint) are present. If missing, either set defaults or exit.
Megatron-LM/tools/checkpoint/loader_base.py:108
↓ 1 callersMethodentry
(self, input_file, rollout_func, reward_func, num_epoch=1)
slime/slime_plugins/rollout_buffer/generator/base_generator.py:267
↓ 1 callersMethoderror_message
(self, metric_name: str)
Megatron-LM/tests/functional_tests/python_test_utils/common.py:47
↓ 1 callersFunctionestimator
Calculates 1 - comb(n - c, k) / comb(n, k).
slime/slime/utils/metric_utils.py:48
↓ 1 callersFunctioneval_multi_choice
Evaluate a multiple choice instance.
Megatron-LM/examples/multimodal/evaluation/mmmu_utils.py:321
↓ 1 callersFunctioneval_only
(args)
slime/eval_only.py:8
↓ 1 callersFunctioneval_open
Evaluate an open question instance
Megatron-LM/examples/multimodal/evaluation/mmmu_utils.py:338
↓ 1 callersFunctioneval_rollout_single_dataset
An example to implement the eval_rollout function for an rule based rm rollout generation. Args: args: the whole args rollout_id:
slime/slime/rollout/sglang_rollout.py:440
↓ 1 callersFunctioneval_tasks
Vision language model text generation for single or batch tasks.
Megatron-LM/examples/multimodal/run_text_generation.py:843
↓ 1 callersFunctionevaluate
Evaluation.
Megatron-LM/megatron/training/training.py:2685
↓ 1 callersFunctionevaluate
Batch evaluation for multiple choice and open questions.
Megatron-LM/examples/multimodal/evaluation/mmmu_utils.py:367
↓ 1 callersMethodevaluate
(self, lease_id: str)
terminal-rl/env_client.py:89
↓ 1 callersMethodevaluate
(self, run_lease_id: str)
terminal-rl/remote/pool_server.py:271
↓ 1 callersMethodevaluate
(self, lease_id: str)
gui-rl/env_pool_server.py:244
↓ 1 callersMethodevaluate
Evaluate whether the task is successfully completed.
gui-rl/desktop_env/desktop_env.py:425
↓ 1 callersMethodevaluate
Apply patch + run eval script. Returns {ok, resolved, ...}.
swe-rl/swe_env_client.py:104
↓ 1 callersMethodevaluate
(self, lease_id: str, patch: str, eval_script: str, cwd: str = "/testbed", timeout: int = 360
swe-rl/server/swe_env_pool_server.py:163
↓ 1 callersFunctionevaluate_batch
Score a batch of (question, response) pairs in parallel. Returns list of scores. Failed evaluations are -1.0.
openclaw-rl/oel/eval/personalization_evaluator.py:179
↓ 1 callersFunctionevaluate_ems
(prediction_file, ground_truth_file, dev_num=3000)
Megatron-LM/tools/retro/text_generation/evaluate.py:122
↓ 1 callersFunctionevaluate_equation
Safely evaluate the arithmetic equation using eval() with precautions.
Megatron-LM/examples/rl/environments/countdown/countdown.py:55
↓ 1 callersFunctionevaluate_f1
(ground_truth_file, prediction_file, reduced_test_only=False)
Megatron-LM/tools/retro/text_generation/evaluate.py:173
↓ 1 callersMethodevaluation
( self, prompt: Any, golden: dict | None, request: EvaluationRequest )
Megatron-LM/megatron/rl/agent/pass_at_evaluation_agent.py:58
↓ 1 callersMethodevaluation_prompts
Get evaluation prompts for the agent. This method should be overridden by subclasses.
Megatron-LM/megatron/rl/agent/reward_only_agent.py:56
↓ 1 callersFunctionexact_match_score
(prediction, ground_truth)
Megatron-LM/tools/retro/text_generation/evaluate.py:114
↓ 1 callersFunctionexchange_loaded_objects_gather_object
Exchange the objects loaded by different ranks with a simple all_gather_object call. Args: loaded_objects (Dict[_ShardId, Any]): mapping
Megatron-LM/megatron/core/dist_checkpointing/exchange_utils.py:424
↓ 1 callersMethodexec
Execute a command in the container. Returns {ok, returncode, output}.
swe-rl/swe_env_client.py:73
↓ 1 callersMethodexec
(self, lease_id: str, command: str, cwd: str = "/testbed", timeout: int = 180, env: dict | None =
swe-rl/server/swe_env_pool_server.py:144
↓ 1 callersMethodexec_tool
( self, lease_id: str, tool_name: str, arguments: dict[str, Any] )
terminal-rl/env_client.py:74
↓ 1 callersMethodexec_tool
( self, run_lease_id: str, tool_name: str, arguments: dict[str, Any] | None = None )
terminal-rl/remote/pool_server.py:260
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen2.5_0.5B_gsm8k_short.py:17
↓ 1 callersFunctionexecute
Run training with MTP enabled and very short output length to cause truncation.
slime/tests/test_mimo_7B_mtp_only_grad.py:37
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_0.6B_fsdp_colocated_2xGPU.py:13
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_30B_A3B_r3.py:26
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_30B_A3B.py:26
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_4B_ppo.py:23
↓ 1 callersFunctionexecute
()
slime/tests/test_quick_start_glm4_9B.py:21
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen35_4b_fsdp.py:49
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen2.5_0.5B_gsm8k_async_short.py:17
↓ 1 callersFunctionexecute
()
slime/tests/test_external_rollout.py:22
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen2.5_0.5B_gsm8k.py:19
↓ 1 callersFunctionexecute
(args)
slime/tests/test_qwen3_4B_fsdp_true_on_policy.py:21
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen2.5_0.5B_gsm8k_async.py:18
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_0.6B_parallel_check.py:24
↓ 1 callersFunctionexecute
()
slime/tests/test_moonlight_16B_A3B_r3.py:23
↓ 1 callersFunctionexecute
()
slime/tests/test_moonlight_16B_A3B.py:23
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_0.6B_megatron_fsdp_align.py:26
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_0.6B_fsdp_distributed.py:16
↓ 1 callersFunctionexecute
()
slime/tests/test_qwen3_vl_4B_fsdp.py:17
↓ 1 callersFunctionexecute
()
slime/examples/true_on_policy/run_simple.py:21
↓ 1 callersFunctionexecute
()
slime/examples/geo3k_vlm_multi_turn/run_geo3k_vlm_multi_turn.py:42
↓ 1 callersFunctionexecute
()
slime/examples/true_on_policy_vlm/run_simple.py:21
↓ 1 callersMethodexecute_action
(self, action: dict)
swe-rl/mini-swe-agent/src/minisweagent/agents/default.py:114
↓ 1 callersMethodexecute_code
Execute Python code in sandbox with safety checks
toolcall-rl/tool_sandbox.py:193
↓ 1 callersFunctionexecute_predictions
Execute predictions and return results
toolcall-rl/generate_with_retool.py:533
↓ 1 callersFunctionexecute_predictions
(prediction: str)
slime/examples/search-r1/generate_with_search.py:124
↓ 1 callersMethodexecute_sync
Helper to synchronously execute the request. This logic is equivalent to what should happen in case of the async call.
Megatron-LM/megatron/core/dist_checkpointing/strategies/async_utils.py:77
↓ 1 callersMethodexecute_tool
Execute a tool call with the given arguments
toolcall-rl/tool_sandbox.py:337
↓ 1 callersFunctionexpected_rel_bound
Bound ~ k * (C ** (L + 1 - l)) * eps_mch, with 1-based layer index l. - L is hard-coded default to 32 per your request. - C is 'close to
Megatron-LM/tests/functional_tests/python_test_utils/test_optimizer_grads_match.py:59
↓ 1 callersFunctionexpert_param_local_key
Get the module parameter corresponding to the key. Args: key: The parameter key to process. num_experts: Total number of experts
Megatron-LM/megatron/core/transformer/fsdp_dtensor_checkpoint.py:163
↓ 1 callersFunctionextract_answer
(text)
Megatron-LM/examples/multimodal/evaluation/evaluate_mmmu.py:37
↓ 1 callersFunctionextract_answer
Extract answer.
Megatron-LM/examples/multimodal/evaluation/evaluate_mathvista.py:60
↓ 1 callersFunctionextract_content
(response)
openclaw-rl/oel/eval/select_hard_problems.py:92
← previousnext →3,801–3,900 of 10,561, ranked by callers