MCPcopy Create free account

hub / github.com/OpenBMB/ToolBench / functions

Functions374 in github.com/OpenBMB/ToolBench

⨍ Functions 374◇ Types & classes 56↳ Endpoints 2
↓ 17 callersMethodadd_to_queue
(self, method_name: str, block_id, **kwargs: Any)
toolbench/inference/callbacks/ServerEventCallback.py:13
↓ 15 callersFunctionstandardize
(string)
toolbench/utils.py:44
↓ 12 callersMethodadd_node
(self,node:ExecutionNode)
toolbench/tooleval/evaluation/dataclass.py:197
↓ 8 callersFunctionchange_name
(name)
toolbench/utils.py:66
↓ 7 callersMethodcheck_success
Returns 1 if successful, otherwise returns 0
toolbench/inference/Downstream_tasks/base_env.py:29
↓ 7 callersMethodfunction_call
(self, func_name, func_args:Dict, *,
toolbench/tooleval/evaluators/registered_cls/tooleval.py:70
↓ 7 callersMethodget_adjacent_node
(self,node:Union[ExecutionNode,GID])
toolbench/tooleval/evaluation/dataclass.py:235
↓ 6 callersMethodget_depth
(self)
toolbench/inference/Tree/Tree.py:66
↓ 6 callersMethodprint
(self,process_id = 0)
toolbench/inference/Tree/Tree.py:88
↓ 6 callersFunctionrank0_print
(*args)
toolbench/train/train.py:76
↓ 6 callersMethodreduce_graph_to_sequence
(self)
toolbench/tooleval/evaluation/dataclass.py:94
↓ 5 callersMethodappend_message
Append a new message.
toolbench/tool_conversation.py:139
↓ 5 callersMethodcheck_task_solvable
(self, task_description:Dict, has_been_solved=False,
toolbench/tooleval/evaluators/registered_cls/rtl.py:99
↓ 5 callersFunctionget_conv_template
Get a conversation template.
toolbench/tool_conversation.py:200
↓ 5 callersFunctionget_conversation_template
(model_path: str)
toolbench/model/model_adapter.py:177
↓ 5 callersFunctionload_registered_automatic_evaluator
(config:dict={},evaluator_name=None,evaluators_cfg_path=None)
toolbench/tooleval/evaluators/__init__.py:7
↓ 4 callersMethodannotate_preference
Annotate and return the index of the preferred answer. For given query, available tools, and two answers, return the index of the pre
toolbench/tooleval/evaluators/registered_cls/base.py:41
↓ 4 callersMethodconvert_to_dict
(self)
toolbench/tooleval/evaluation/dataclass.py:70
↓ 4 callersMethodcopy
(self)
toolbench/tool_conversation.py:165
↓ 4 callersFunctionget_most_preferred
(d:list)
toolbench/tooleval/evaluators_comparison.py:18
↓ 4 callersFunctionget_steps
(example)
toolbench/tooleval/utils.py:21
↓ 4 callersFunctionprocess_system_message
(system_message, functions)
toolbench/utils.py:9
↓ 4 callersFunctionreact_parser
(string)
toolbench/inference/utils.py:34
↓ 4 callersFunctionregister_conv_template
Register a new conversation template.
toolbench/tool_conversation.py:193
↓ 4 callersFunctionregister_model_adapter
Register a model adapter.
toolbench/model/model_adapter.py:49
↓ 3 callersMethodadd_edge
(self,from_node:Union[ExecutionNode,GID],to_node:Union[ExecutionNode,GID],edge:DirectedEdge=None)
toolbench/tooleval/evaluation/dataclass.py:203
↓ 3 callersFunctionagreement_score
(x,ref:list)
toolbench/tooleval/evaluators_comparison.py:28
↓ 3 callersMethodchange_messages
(self,messages)
toolbench/inference/LLM/davinci_model.py:49
↓ 3 callersMethodcheck_is_solved
(self, task_description:Dict, answer:Dict[Any,Any],
toolbench/tooleval/evaluators/registered_cls/rtl.py:61
↓ 3 callersMethodget_chain_result_from_this_node
Returns chained results, starting from this node up to the root node
toolbench/inference/Tree/Tree.py:171
↓ 3 callersMethodget_former_trice_from_this_node
Return path description from end_node -> self Does not contain end_node, never contains root node
toolbench/inference/Tree/Tree.py:182
↓ 3 callersMethodget_train_messages_from_this_node
Returns chained results, starting from this node up to the root node
toolbench/inference/Tree/Tree.py:137
↓ 3 callersMethodparse
(self,functions,process_id,**args)
toolbench/inference/LLM/davinci_model.py:73
↓ 3 callersMethodprediction
(self, prompt: str, stop: Optional[List[str]] = None)
toolbench/inference/LLM/davinci_model.py:21
↓ 3 callersMethodreturn_output
Return output.
toolbench/inference/utils.py:237
↓ 3 callersMethodrun
(self)
toolbench/inference/Downstream_tasks/rapidapi.py:531
↓ 3 callersMethodselect_better_answer
(self, task_description:Dict, task_status:TaskStatus,
toolbench/tooleval/evaluators/registered_cls/rtl.py:168
↓ 3 callersMethodset_init_node
(self,node:Union[GID,ExecutionNode])
toolbench/tooleval/evaluation/dataclass.py:181
↓ 2 callersFunctionchat_completion_request
(key, messages, functions=None,function_call=None,key_pos=None, model="gpt-3.5-turbo-16k-0613",stop=None,proce
toolbench/inference/LLM/chatgpt_function_model.py:10
↓ 2 callersMethodcompare_answer_details
(self,answer:List)
toolbench/tooleval/evaluators/registered_cls/tooleval.py:125
↓ 2 callersFunctioncontain
(candidate_list, white_list)
toolbench/inference/Downstream_tasks/rapidapi.py:45
↓ 2 callersMethodfetch_api_json
(self, query_json)
toolbench/inference/Downstream_tasks/rapidapi.py:157
↓ 2 callersFunctiongenerate_init_message_node
(eg:ExecutionGraph,functions,query)
toolbench/tooleval/convert_to_answer_format.py:16
↓ 2 callersFunctiongenerate_init_message_node
(eg:ExecutionGraph,functions,query)
toolbench/tooleval/utils.py:45
↓ 2 callersMethodget_backbone_model
(self)
toolbench/inference/Downstream_tasks/rapidapi.py:400
↓ 2 callersFunctionget_model_adapter
Get a model adapter for a model_path.
toolbench/model/model_adapter.py:55
↓ 2 callersFunctionget_pass_rate_results
(filename: str)
toolbench/tooleval/eval_preference.py:51
↓ 2 callersMethodget_prompt
Get the prompt for generation.
toolbench/tool_conversation.py:47
↓ 2 callersFunctionget_rapidapi_response
(input_dict: dict, api_customization: bool=False, tools_root: str="data.toolenv.tools", schema_root: str="data
toolbench/inference/server.py:129
↓ 2 callersMethodget_retriever
(self)
toolbench/inference/Downstream_tasks/rapidapi.py:414
↓ 2 callersFunctionget_white_list
(tool_root_dir)
toolbench/inference/Downstream_tasks/rapidapi.py:27
↓ 2 callersMethodmake_finish
Recursively marked as finish, until the above inter_val nodes of action_input type (including yourself)
toolbench/inference/Tree/Tree.py:126
↓ 2 callersFunctionmake_supervised_data_module
Make dataset and collator for supervised fine-tuning.
toolbench/train/train.py:225
↓ 2 callersMethodmatch
(self, model_path: str)
toolbench/model/model_adapter.py:31
↓ 2 callersMethodpop_node
(self,node:Union[ExecutionNode,GID])
toolbench/tooleval/evaluation/dataclass.py:221
↓ 2 callersMethodprediction
(self, prompt: str, stop: Optional[List[str]] = None)
toolbench/inference/LLM/llama_model.py:35
↓ 2 callersMethodprediction
(self, prompt: str, stop: Optional[List[str]] = None)
toolbench/inference/LLM/tool_llama_lora_model.py:53
↓ 2 callersMethodprediction
(self, prompt: str, stop: Optional[List[str]] = None)
toolbench/inference/LLM/tool_llama_model.py:43
↓ 2 callersFunctionpreprocess
( sources, tokenizer: transformers.PreTrainedTokenizer, template: str="tool-llama" )
toolbench/train/train.py:90
↓ 2 callersFunctionprint_and_save_leaderboard
(leaderboard)
toolbench/tooleval/eval_and_update_leaderboard.py:79
↓ 2 callersFunctionprocess_data
(data, pairs)
preprocess/preprocess_retriever_data.py:70
↓ 2 callersFunctionprocess_invalid_data
(method,data_dict)
toolbench/tooleval/convert_to_answer_format.py:74
↓ 2 callersFunctionprocess_retrieval_ducoment
(documents_df)
toolbench/utils.py:112
↓ 2 callersFunctionprocess_valid_data
(method,answer_generation)
toolbench/tooleval/convert_to_answer_format.py:27
↓ 2 callersFunctionreplace_llama_attn_with_flash_attn
()
toolbench/train/llama_flash_attn_monkey_patch.py:73
↓ 2 callersFunctionreplace_llama_with_condense
(ratio)
toolbench/train/llama_condense_monkey_patch.py:43
↓ 2 callersMethodrequest
(self,messages,**kwargs)
toolbench/tooleval/evaluators/registered_cls/utils.py:49
↓ 2 callersMethodrun_single_task
(self, method, backbone_model, query_id, data_dict, args, output_dir_path, tool_des, retriever=None, process_i
toolbench/inference/Downstream_tasks/rapidapi.py:484
↓ 2 callersFunctionsplit_files
(model_path, tmp_path, split_size)
toolbench/model/apply_delta.py:25
↓ 2 callersMethodstart
single_chain_max_step: The maximum depth of the tree tree_beam_size: How many children nodes for one node are generated per layer
toolbench/inference/Algorithms/DFS.py:94
↓ 2 callersMethodstep
Perform an interaction in natural language mode return value (output str, status code)
toolbench/inference/Downstream_tasks/base_env.py:22
↓ 2 callersMethodto_json
(self, use_messages=False)
toolbench/inference/Tree/Tree.py:210
↓ 2 callersMethodto_json_recursive
(self,use_messages=False)
toolbench/inference/Tree/Tree.py:13
↓ 2 callersFunctiontrain
()
toolbench/train/train.py:251
↓ 1 callersFunctionDFS
(root)
toolbench/tooleval/convert_to_answer_format.py:108
↓ 1 callersFunctionDFS
(root)
toolbench/tooleval/utils.py:133
↓ 1 callersMethodDFS
Returns the number of grids to go back. When a child node of a node generates a final answer or give up, it should go back a few more grids In
toolbench/inference/Algorithms/DFS.py:120
↓ 1 callersMethod__init__
(self, fn_completions: Callable[[Dict,List[Dict]],int] = None, *args,
toolbench/tooleval/evaluators/registered_cls/base.py:36
↓ 1 callersMethod__init__
(self, cfg_path: str = None, )
toolbench/tooleval/evaluators/registered_cls/tooleval.py:17
↓ 1 callersMethod__init__
(self, args, add_retrieval=False, process_id=0, server=False)
toolbench/inference/Downstream_tasks/rapidapi.py:392
↓ 1 callersMethod__init__
(self, raw_data, tokenizer: transformers.PreTrainedTokenizer, template="tool-llama")
toolbench/train/train.py:172
↓ 1 callersMethod_forward
(self,query:str,tools:List[Dict])
toolbench/tooleval/evaluation/usereval.py:52
↓ 1 callersMethod_step
Need to return an observation string and status code: 0 means normal response 1 means there is no corresponding api name
toolbench/inference/Downstream_tasks/rapidapi.py:286
↓ 1 callersMethodapi_json_to_openai_json
(self, api_json,standard_tool_name)
toolbench/inference/Downstream_tasks/rapidapi.py:185
↓ 1 callersFunctionappend_list
(instances_list: list)
preprocess/preprocess_toolllama_data.py:29
↓ 1 callersFunctionapply_compressed_weight
(module, compressed_state_dict, target_device, prefix="")
toolbench/model/compression.py:61
↓ 1 callersFunctionapply_delta
(base_model_path, target_model_path, delta_path)
toolbench/model/apply_delta.py:125
↓ 1 callersFunctionapply_delta_low_cpu_mem
(base_model_path, target_model_path, delta_path)
toolbench/model/apply_delta.py:70
↓ 1 callersMethodbuild_corpus_embeddings
(self)
toolbench/inference/LLM/retriever.py:30
↓ 1 callersMethodbuild_retrieval_corpus
(self)
toolbench/inference/LLM/retriever.py:17
↓ 1 callersMethodbuild_retrieval_embedder
(self)
toolbench/inference/LLM/retriever.py:25
↓ 1 callersMethodbuild_tool_description
(self, data_dict)
toolbench/inference/Downstream_tasks/rapidapi.py:131
↓ 1 callersFunctioncalculate_evaluator_performance
(evaluator_preference,human_preference)
toolbench/tooleval/evaluators_comparison.py:101
↓ 1 callersFunctioncalculate_human_performance
()
toolbench/tooleval/evaluators_comparison.py:81
↓ 1 callersMethodchange_messages
(self,messages)
toolbench/inference/LLM/llama_model.py:54
↓ 1 callersMethodchange_messages
(self,messages)
toolbench/inference/LLM/tool_llama_lora_model.py:72
↓ 1 callersMethodchange_messages
(self,messages)
toolbench/inference/LLM/chatgpt_function_model.py:58
↓ 1 callersMethodchange_messages
(self,messages)
toolbench/inference/LLM/tool_llama_model.py:63
↓ 1 callersMethodcheck_has_hallucination
(self,available_tools:List[Dict],answer:Dict[Any,Any])
toolbench/tooleval/evaluators/registered_cls/rtl.py:34
↓ 1 callersMethodcheck_identity_answers
(self, answers:List[Dict[Any,Any]], )
toolbench/tooleval/evaluators/registered_cls/rtl.py:156
next →1–100 of 374, ranked by callers