MCPcopy Create free account

hub / github.com/ZJU-REAL/SkillZero / functions

Functions3,531 in github.com/ZJU-REAL/SkillZero

⨍ Functions 3,531◇ Types & classes 522↳ Endpoints 90
↓ 3 callersFunctionfind_latest_ckpt_path
Return the most recent checkpoint directory based on a tracker file. Args: path (str): Base directory containing the checkpoint trac
verl/utils/checkpoint/checkpoint_manager.py:135
↓ 3 callersMethodfinish_of_episode
(self, episode_no, batch_size)
agent_system/environments/env_package/alfworld/alfworld/agents/agent/base_agent.py:353
↓ 3 callersMethodfit
The training loop of PPO. The driver process only need to call the compute functions of the worker group through RPC to const
verl/trainer/ppo/ray_trainer.py:1237
↓ 3 callersMethodfix_and_comma_in_the_end
(self, desc)
agent_system/environments/env_package/alfworld/alfworld/agents/controller/base.py:91
↓ 3 callersFunctionfree_NormEffect
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/memory.c:1047
↓ 3 callersMethodfrom_detached
Create a worker group from existing detached workers. Args: name_prefix: Prefix for worker names worker_names: Names
verl/single_controller/ray/base.py:407
↓ 3 callersFunctionfsdp2_clip_grad_norm_
torch.nn.utils.clip_grad_norm_ cann't run on cpu parameter DTensor
verl/utils/fsdp_utils.py:447
↓ 3 callersMethodget_active_skill_names
Return names of currently active skill files (for logging).
agent_system/environments/curriculum_skill_manager.py:186
↓ 3 callersFunctionget_aggregated
Get the aggregated hidden state of the encoder. B x D
agent_system/environments/env_package/webshop/webshop/baseline_models/models/modules.py:18
↓ 3 callersMethodget_available_actions
Returns list of available actions at the current step
agent_system/environments/env_package/webshop/webshop/web_agent_site/envs/web_agent_site_env.py:93
↓ 3 callersMethodget_batch
(self, batch_size, multi_step=1)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/memory.py:100
↓ 3 callersMethodget_batch_of_sequences
(self, batch_size, sample_history_length)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/memory.py:216
↓ 3 callersFunctionget_data
(split)
agent_system/environments/env_package/webshop/webshop/baseline_models/train_search_il.py:34
↓ 3 callersFunctionget_default_kwargs_for_model_parallel_config
()
verl/utils/megatron/tensor_parallel.py:35
↓ 3 callersMethodget_env_paths
(self)
agent_system/environments/env_package/alfworld/alfworld/agents/environment/alfred_thor_env.py:237
↓ 3 callersFunctionget_fsdp_state_ctx
(model, state_type, state_cfg, optim_cfg)
verl/utils/fsdp_utils.py:387
↓ 3 callersFunctionget_image
(card_name)
agent_system/environments/env_package/gym_cards/gym-cards/gym_cards/envs/blackjack.py:12
↓ 3 callersMethodget_key
Generate a fast content key for a segment. Args: segment_text: The normalized segment text
agentocr/ocrtool.py:64
↓ 3 callersFunctionget_mcore_forward_fn
Get the forward function for given model architecture.
verl/models/mcore/registry.py:235
↓ 3 callersFunctionget_megatron_optimizer
( model, config: OptimizerConfig, no_weight_decay_cond=None, scale_lr_cond=None, lr_mult=1
verl/utils/megatron/optimizer.py:20
↓ 3 callersFunctionget_model_checkpoint_path
(checkpoint_path)
verl/utils/megatron_utils.py:431
↓ 3 callersMethodget_obj_cls_from_metadata
(self, name)
agent_system/environments/env_package/alfworld/alfworld/agents/controller/oracle.py:38
↓ 3 callersFunctionget_reward
Get cumulative reward score for purchased product and goal
agent_system/environments/env_package/webshop/webshop/web_agent_site/engine/goal.py:228
↓ 3 callersFunctionget_rollout_config
(max_response_length, max_prompt_length, dtype, tensor_parallel_size, tool_config_path)
tests/workers/rollout/utils_sglang.py:123
↓ 3 callersFunctionget_seqlen_balanced_partitions
Calculates partitions of indices from seqlen_list such that the sum of sequence lengths in each partition is balanced. Uses the Karmarkar-Kar
verl/utils/seqlen_balancing.py:143
↓ 3 callersFunctionget_sharding_strategy
(device_mesh)
verl/workers/fsdp_workers.py:88
↓ 3 callersMethodget_some_visible_obj_of_name
(self, name)
agent_system/environments/env_package/alfworld/alfworld/gen/game_states/game_state_base.py:370
↓ 3 callersMethodget_state
(self)
tests/e2e/envs/digit_completion/task.py:59
↓ 3 callersMethodget_stats
Get cache statistics.
agentocr/ocrtool.py:116
↓ 3 callersMethodget_task_and_obs
(self, observation_strings)
agent_system/environments/env_package/alfworld/alfworld/agents/agent/base_agent.py:413
↓ 3 callersMethodget_teleport_action
(self, action)
agent_system/environments/env_package/alfworld/alfworld/gen/game_states/planned_game_state.py:378
↓ 3 callersFunctionget_templated_task_desc
(traj_data)
agent_system/environments/env_package/alfworld/alfworld/agents/utils/misc.py:79
↓ 3 callersFunctionget_ulysses_sequence_parallel_rank
Get ulysses sequence parallel rank.
verl/utils/ulysses.py:54
↓ 3 callersFunctionget_version
(pkg)
verl/workers/rollout/vllm_rollout/__init__.py:20
↓ 3 callersMethodget_vocab
(self)
tests/e2e/envs/digit_completion/tokenizer.py:86
↓ 3 callersFunctionhash_bfs_node
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/search.c:1706
↓ 3 callersFunctionhf_to_mcore_config
Convert huggingface PretrainedConfig to mcore TransformerConfig. Args: hf_config: The huggingface PretrainedConfig. dtype: The dt
verl/models/mcore/registry.py:179
↓ 3 callersFunctioninfo
(*args)
agent_system/environments/env_package/webshop/webshop/baseline_models/logger.py:266
↓ 3 callersFunctioninit_async_rollout_manager
(config: DictConfig, scheduler_kwargs: Dict[str, Any] = None)
tests/workers/rollout/async_rollout_utils.py:27
↓ 3 callersFunctioninit_megatron_optim_config
(optim_config: Dict)
verl/utils/megatron_utils.py:201
↓ 3 callersMethodinit_workers
Initialize distributed training workers using Ray backend. Creates: 1. Ray resource pools from configuration 2. Worker groups
verl/trainer/ppo/ray_trainer.py:1058
↓ 3 callersFunctioninitialize_fixpoint
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/relax.c:789
↓ 3 callersFunctioninstantiate_wff
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_hard.c:662
↓ 3 callersFunctioninstantiated_fact_adress
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_hard.c:979
↓ 3 callersFunctionintroduce_benefits_and_enforcements
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/relax.c:2329
↓ 3 callersFunctionintroduce_pc_goals
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/relax.c:2516
↓ 3 callersFunctionis_dist_avail_and_initialized
()
agent_system/environments/env_package/alfworld/alfworld/agents/detector/utils.py:268
↓ 3 callersFunctionis_natural
(hand)
agent_system/environments/env_package/gym_cards/gym-cards/gym_cards/envs/blackjack.py:75
↓ 3 callersMethodis_padding_enabled
Check if padding is enabled for the DataProto. Returns: bool: True if padding is enabled, False otherwise.
verl/protocol.py:720
↓ 3 callersFunctionkl_penalty
Compute KL divergence given logprob and ref_logprob. Copied from https://github.com/huggingface/trl/blob/main/trl/trainer/ppo_trainer.py#L1104
verl/trainer/ppo/core_algos.py:615
↓ 3 callersMethodlayer_dropout
(self, inputs, residual, dropout)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:1088
↓ 3 callersMethodlayer_dropout
(self, inputs, residual, dropout)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:1145
↓ 3 callersMethodlayer_dropout
(self, inputs, residual, dropout)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:1205
↓ 3 callersMethodload_checkpoint
(self, local_path, hdfs_path=None, del_local_after_load=True)
verl/workers/fsdp_workers.py:1119
↓ 3 callersFunctionload_docs
(corpus, doc_idxs)
examples/search/retriever/retrieval_server.py:30
↓ 3 callersFunctionload_products
(filepath, attrpath, num_products=None, human_goals=True)
agent_system/environments/env_package/webshop/webshop/web_agent_site/engine/engine.py:230
↓ 3 callersFunctionload_reward_manager
(config, tokenizer, num_examine, **reward_kwargs)
verl/trainer/ppo/reward.py:60
↓ 3 callersFunctionload_skill_file
Load a skill file with section markers like '### SECTION_NAME ###'.
agent_system/environments/env_manager.py:41
↓ 3 callersFunctionload_tensor_to_gpu
(tensor)
verl/utils/megatron_utils.py:357
↓ 3 callersMethodlog
(self, func, *args, **kwargs)
verl/utils/debug/performance.py:84
↓ 3 callersFunctionlogical_state_sum
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/search.c:1848
↓ 3 callersFunctionmake_Wff
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_pre.c:809
↓ 3 callersFunctionmake_either_ty_in_pl
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/parse.c:742
↓ 3 callersFunctionmake_map_fn
(split)
examples/data_preprocess/hellaswag.py:54
↓ 3 callersMethodmerge
(self, other)
verl/utils/seqlen_balancing.py:34
↓ 3 callersFunctionmerge_next_step_ANDs_and_ORs_in_wff
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_pre.c:3780
↓ 3 callersFunctionnew_Action
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/memory.c:760
↓ 3 callersFunctionnew_EhcNode
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/memory.c:796
↓ 3 callersFunctionnew_NumericEffect
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/memory.c:412
↓ 3 callersFunctionnormalize_final_answer
Normalize a final answer to a quantitative reasoning question. Args: final_answer: The answer string to normalize Returns: N
verl/utils/reward_score/math_dapo.py:125
↓ 3 callersFunctionnormalize_tyl_in_pl
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/parse.c:786
↓ 3 callersMethodobject_id_to_cls
(self, object_hash, special_char=' ')
agent_system/environments/env_package/alfworld/alfworld/agents/expert/handcoded_expert.py:59
↓ 3 callersFunctionoffload_tensor_to_cpu
(tensor)
verl/utils/megatron_utils.py:317
↓ 3 callersMethodpad_and_stack
(self, obs, max_feat_len=None)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/layers.py:1612
↓ 3 callersFunctionparse_action
Parse action string to action name and its arguments.
agent_system/environments/env_package/webshop/webshop/web_agent_site/engine/engine.py:117
↓ 3 callersFunctionper_tensor_generator
(actor_module, model_config, weight_converter, transformer_config, layer_name_mapping, convert_qkv_gate_up_by_
verl/utils/megatron_utils.py:727
↓ 3 callersFunctionposition_in_constants_table
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_pre.c:375
↓ 3 callersFunctionpreproc
(s)
agent_system/environments/env_package/alfworld/alfworld/agents/modules/generic.py:115
↓ 3 callersFunctionprint_Action_name
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/output.c:1098
↓ 3 callersFunctionprint_fixpoint_result
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/relax.c:1459
↓ 3 callersMethodprint_frame
(self, recep, loc)
agent_system/environments/env_package/alfworld/alfworld/agents/controller/mrcnn.py:191
↓ 3 callersMethodprint_frame
(self, recep, loc)
agent_system/environments/env_package/alfworld/alfworld/agents/controller/oracle.py:125
↓ 3 callersFunctionprint_successes
(succ_traj)
agent_system/environments/env_package/alfworld/alfworld/gen/scripts/generate_trajectories.py:376
↓ 3 callersFunctionprocess_validation_metrics
Process validation metrics into a structured format with statistical analysis. This function organizes validation metrics by data source
verl/trainer/ppo/metric_utils.py:374
↓ 3 callersMethodpush_first_sight
(self, stuff)
agent_system/environments/env_package/alfworld/alfworld/agents/agent/base_agent.py:56
↓ 3 callersFunctionqwen2_vl_forward
( self: "Qwen2VLForConditionalGeneration", input_ids: torch.LongTensor, attention_mask: Optional[t
verl/models/transformers/qwen2_vl.py:431
↓ 3 callersMethodreceive
Map action to the corresponding page
agent_system/environments/env_package/webshop/webshop/web_agent_site/envs/web_agent_text_env.py:519
↓ 3 callersMethodreduce
Returns result of applying `self.operation` to a contiguous subsequence of the array. self.operation(arr[start], operati
agent_system/environments/env_package/alfworld/alfworld/agents/modules/segment_tree.py:46
↓ 3 callersMethodrelease
Release the tool instance. Args: instance_id: The instance id of the tool.
verl/tools/base_tool.py:80
↓ 3 callersFunctionremove_unused_vars_in_wff
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_pre.c:1417
↓ 3 callersMethodrender
Implement this if visualization is needed.
agent_system/environments/env_package/appworld/envs.py:234
↓ 3 callersMethodrender
(self, mode)
agent_system/environments/env_package/sokoban/sokoban/env.py:107
↓ 3 callersFunctionreplace_not_p_with_n_in_wff
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/inst_pre.c:2952
↓ 3 callersMethodrl_forward
(self, state_batch, act_batch, value=False, q=False, act=False)
agent_system/environments/env_package/webshop/webshop/baseline_models/models/bert.py:86
↓ 3 callersFunctionsame_lnfs
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/expressions.c:1782
↓ 3 callersFunctionsame_state
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/search.c:792
↓ 3 callersFunctionselect_op
agent_system/environments/env_package/alfworld/alfworld/gen/ff_planner/relax.c:2293
↓ 3 callersMethodset_input_tensor
Set input tensor to be used instead of forward()'s input. When doing pipeline parallelism the input from the previous stage comes fro
verl/models/qwen2/megatron/modeling_qwen2_megatron.py:453
↓ 3 callersMethodset_level
(self, level)
agent_system/environments/env_package/webshop/webshop/baseline_models/logger.py:374
↓ 3 callersFunctionset_random_seed
(seed)
verl/workers/megatron_workers.py:52
← previousnext →501–600 of 3,531, ranked by callers