Functions — github.com/SqueezeAILab/LLMCompiler

hub / github.com/SqueezeAILab/LLMCompiler / functions

Functions245 in github.com/SqueezeAILab/LLMCompiler

⨍ Functions 245◇ Types & classes 38

↓ 1 callersMethodingest_token

(self, token: str)

src/llm_compiler/planner.py:127

↓ 1 callersFunctioninitialize_react_agent_executor

Load an agent executor given tools and LLM. Simplified version of langchain.agents.initialize.initialize_agent

src/react/base.py:44

↓ 1 callersFunctionis_number

(s)

src/utils/evaluation_utils.py:55

↓ 1 callersFunctionlower

(text)

src/utils/evaluation_utils.py:19

↓ 1 callersFunctionmain

()

run_llm_compiler.py:129

↓ 1 callersMethodparse

(self, text: str)

src/llm_compiler/output_parser.py:32

↓ 1 callersMethodplan

( self, inputs: dict, is_replan: bool, callbacks: Callbacks = None, **kwargs: Any )

src/llm_compiler/planner.py:256

↓ 1 callersMethodplan

Given input, decided what to do. Args: intermediate_steps: Steps the LLM has taken to date, along with observatio

src/agents/agent.py:99

↓ 1 callersMethodprep_prompts

Prepare prompts from inputs.

src/chains/llm_chain.py:172

↓ 1 callersFunctionremove_articles

(text)

src/utils/evaluation_utils.py:9

↓ 1 callersFunctionremove_punc

(text)

src/utils/evaluation_utils.py:15

↓ 1 callersFunctionreplace_min_max_functions

(expression)

src/chains/llm_math_chain.py:116

↓ 1 callersMethodreset

(self)

src/callbacks/callbacks.py:29

↓ 1 callersMethodreset_all_stats

(self)

src/llm_compiler/llm_compiler.py:128

↓ 1 callersFunctionrun_llm_math_chain_factory

(llm_math_chain)

configs/parallelqa/tools.py:30

↓ 1 callersFunctionrun_llm_math_chain_factory

(llm_math_chain)

configs/parallelqa_react/tools.py:7

↓ 1 callersMethodsave

Raise error - saving not supported for Agent Executors.

src/executors/agent_executor.py:147

↓ 1 callersMethodschedule

Run all tasks in self.tasks in parallel, respecting dependencies.

src/llm_compiler/task_fetching_unit.py:124

↓ 1 callersFunctionwhite_space_fix

(text)

src/utils/evaluation_utils.py:12

Method__call__

(self)

src/llm_compiler/task_fetching_unit.py:53

Method__call__

Execute the chain. Args: inputs: Dictionary of inputs, or single input if chain expects only one param. Should co

src/chains/chain.py:239

Method__init__

Initialize tool.

src/tools/base.py:190

Method__init__

(self)

src/utils/logger_utils.py:12

Method__init__

(self, tools: Sequence[Union[Tool, StructuredTool]])

src/llm_compiler/planner.py:93

Method__init__

( self, queue: asyncio.Queue[Optional[str]], tools: Sequence[Union[Tool, StructuredToo

src/llm_compiler/planner.py:151

Method__init__

( self, llm: BaseChatModel, example_prompt: str, example_prompt_replan: str,

src/llm_compiler/planner.py:191

Method__init__

(self, llm: BaseLLM)

src/llm_compiler/llm_compiler.py:25

Method__init__

Args: tools: List of tools to use. max_replans: Maximum number of replans to do. benchmark: Whether to co

src/llm_compiler/llm_compiler.py:50

Method__init__

(self, tools: Sequence[Union[Tool, StructuredTool]], **kwargs)

src/llm_compiler/output_parser.py:28

Method__init__

(self)

src/llm_compiler/task_fetching_unit.py:87

Method__init__

(self, stream: bool = False)

src/callbacks/callbacks.py:48

Method__init__

Check that wikipedia package is installed.

src/docstore/wikipedia.py:25

Method__init__

Initialize with a docstore, and set initial document to None.

src/docstore/wikipedia.py:232

Method_acall

( self, inputs: Dict[str, Any], run_manager: Optional[AsyncCallbackManagerForChainRun]

src/llm_compiler/llm_compiler.py:234

Method_acall

Run text through and get agent response.

src/executors/agent_executor.py:487

Method_acall

( self, inputs: Dict[str, str], run_manager: Optional[AsyncCallbackManagerForChainRun]

src/chains/llm_math_chain.py:270

Method_acall

( self, inputs: Dict[str, Any], run_manager: Optional[AsyncCallbackManagerForChainRun]

src/chains/llm_chain.py:282

Method_agent_type

Return Identifier of agent type.

src/agents/agent.py:56

Method_agent_type

(self)

src/agents/structured_chat_agent.py:143

Method_aperform_agent_action

( agent_action: AgentAction, )

src/executors/agent_executor.py:395

Method_arun

Use the tool asynchronously.

src/tools/base.py:164

Method_arun

Use the tool asynchronously.

src/tools/base.py:280

Method_arun

( self, query: str, run_manager: Optional[AsyncCallbackManagerForToolRun] = None,

src/executors/agent_executor.py:49

Method_arun

Use the tool asynchronously.

src/agents/tools.py:31

Method_call

( self, inputs: Dict[str, Any], run_manager: Optional[CallbackManagerForChainRun] = No

src/llm_compiler/llm_compiler.py:227

Method_call

Run text through and get agent response.

src/executors/agent_executor.py:439

Method_call

( self, inputs: Dict[str, str], run_manager: Optional[CallbackManagerForChainRun] = No

src/chains/llm_math_chain.py:256

Method_call

( self, inputs: Dict[str, Any], run_manager: Optional[CallbackManagerForChainRun] = No

src/chains/llm_chain.py:95

Method_chain_type

(self)

src/chains/chain.py:138

Method_chain_type

(self)

src/chains/llm_math_chain.py:285

Method_chain_type

(self)

src/chains/llm_chain.py:386

Method_construct_scratchpad

( self, intermediate_steps: List[Tuple[AgentAction, str]] )

src/agents/structured_chat_agent.py:44

Method_fix_text

Fix the text.

src/agents/agent.py:78

Method_get_default_output_parser

src/tools/base.py:414

Method_paragraphs

(self)

src/docstore/wikipedia.py:292

Function_partial

(func: Callable[[str], str])

src/tools/base.py:459

Method_run

Use the tool.

src/tools/base.py:144

Method_run

Use the tool.

src/tools/base.py:260

Method_run

( self, query: str, run_manager: Optional[CallbackManagerForToolRun] = None, )

src/executors/agent_executor.py:42

Method_run

Use the tool.

src/agents/tools.py:18

Method_run_output_key

(self)

src/chains/chain.py:436

Method_run_output_key

(self)

src/chains/llm_chain.py:265

src/agents/agent.py:83

Method_stop

(self)

src/agents/structured_chat_agent.py:70

Method_summary

(self)

src/docstore/wikipedia.py:288

Method_to_args_and_kwargs

Convert tool input to pydantic model.

src/tools/base.py:132

Method_type

(self)

src/react/output_parser.py:32

Method_validate_tools

(cls, tools: Sequence[BaseTool])

src/react/base.py:25

Methodaapply_and_parse

Call apply and then parse the results.

src/chains/llm_chain.py:374

Methodainvoke

( self, input: Union[str, Dict], config: Optional[RunnableConfig] = None, **kw

src/tools/base.py:239

Methodainvoke

( self, input: Dict[str, Any], config: Optional[RunnableConfig] = None, **kwar

src/chains/chain.py:79

Methodapply

Call the chain on all inputs in the list.

src/chains/chain.py:645

Methodapply_and_parse

Call apply and then parse the results.

src/chains/llm_chain.py:352

Methodapredict_and_parse

Call apredict and then parse the results.

src/chains/llm_chain.py:338

Methodargs

The tool's input arguments.

src/tools/base.py:124

Methodargs

The tool's input arguments.

src/tools/base.py:256

Methodasearch

Search for a term in the docstore, and if found save.

src/docstore/wikipedia.py:255

Functioncompute_function

(match)

src/chains/llm_math_chain.py:78

Functiondecorator_time_it

Time a function.

src/utils/time_utils.py:19

Methodfrom_function

Create tool from a given function. A classmethod that helps to create a tool from a function. Args: func: The function f

src/tools/base.py:308

Methodfrom_llm_and_tools

Construct an agent from an LLM and tools.

src/agents/structured_chat_agent.py:103

Methodfrom_string

Create LLMChain from LLM and template.

src/chains/llm_chain.py:390

Functiongenerate_tools

(args)

configs/movie_react/tools.py:5

Functiongenerate_tools

(args, model_name)

configs/parallelqa/tools.py:81

Functiongenerate_tools

(args, model_name)

configs/parallelqa_react/tools.py:32

Functiongenerate_tools

(args)

configs/movie/tools.py:5

Functionget_logger

()

src/utils/logger_utils.py:46

Methodget_results

(self, key: str)

src/utils/logger_utils.py:31

Methodget_stats

(self)

src/callbacks/callbacks.py:93

Methodinput_keys

(self)

src/llm_compiler/llm_compiler.py:135

Methodinput_keys

Return the input keys. :meta private:

src/executors/agent_executor.py:178

Methodinput_keys

Keys expected to be in the chain input.

src/chains/chain.py:171

Methodinput_keys

Expect input key. :meta private:

src/chains/llm_math_chain.py:173

Methodinput_keys

Will be whatever keys the prompt expects. :meta private:

src/chains/llm_chain.py:77

Methodinput_keys

Return the input keys. :meta private:

src/agents/agent.py:171

Methodinvoke

( self, input: Dict[str, Any], config: Optional[RunnableConfig] = None, **kwar

src/chains/chain.py:63

Methoditer

Enables iteration over steps taken to reach final output.

src/executors/agent_executor.py:159

Methodlc_serializable

(self)

src/chains/llm_chain.py:53

← previous next →101–200 of 245, ranked by callers