Module core.python_engine

Classes

class PythonEngine (driver: lavague.core.base_driver.BaseDriver,
llm: llama_index.core.base.llms.base.BaseLLM | None = None,
embedding: llama_index.core.base.embeddings.base.BaseEmbedding | None = None,
logger: lavague.core.logger.AgentLogger | None = None,
clean_html: Callable[[str], str] = <function extract>,
ocr_mm_llm: llama_index.core.base.llms.base.BaseLLM | None = None,
ocr_llm: llama_index.core.base.llms.base.BaseLLM | None = None,
display: bool = False,
batch_size: int = 5,
confidence_threshold: float = 0.85,
fallback_threshold: float = 0.85,
temp_screenshots_path='./tmp_screenshots',
n_search_attemps=10)

The PythonEngine is responsible for knowledge retrieval, it extracts information from the webpage and performs RAG to complete the given instruction

Ancestors

• lavague.core.base_engine.BaseEngine
• abc.ABC
• lavague.core.logger.Loggable
• lavague.core.display.Display

Class variables

var batch_size : int

var clean_html : Callable[[str], str]

var confidence_threshold : float

var driver : lavague.core.base_driver.BaseDriver

var embedding : llama_index.core.base.embeddings.base.BaseEmbedding

var fallback_theshold : float

var llm : llama_index.core.base.llms.base.BaseLLM

var logger : lavague.core.logger.AgentLogger

var n_search_attempts : int

var ocr_llm : llama_index.core.base.llms.base.BaseLLM

var ocr_mm_llm : llama_index.core.base.llms.base.BaseLLM

var temp_screenshots_path : str

Static methods

def from_context(context: lavague.core.context.Context,
driver: lavague.core.base_driver.BaseDriver)

Methods

def display_screenshot(self) ‑> None

def execute_instruction(self, instruction: str) ‑> lavague.core.base_engine.ActionResult

def extract_structured_data(self, output: str) ‑> dict | None

def get_screenshots_batch(self) ‑> list[str]

def perform_fallback(self, prompt, instruction) ‑> str

def set_display(self, display: bool)