Memory

Memory in Eclipse

The Memory class in the Eclipse framework provides a persistent storage mechanism for storing and retrieving context, data, and state during interactions with agents in a pipeline. This functionality enables the system to "remember" information across agent actions and queries, which is particularly useful for complex workflows or tasks that span multiple steps.

The Memory class is initialized with a configuration that defines how it interacts with the agent pipeline. It can be used to persist important data, store intermediate results, or maintain context across different stages of the agent's execution.

Memory Configuration Parameters

Attribute	Parameter	Description
LLM Client	llm_client	Configuration for the Large Language Model (LLM) client.
Vector Store (optional)	vector_store	Configuration for the vector store. Defaults to Chroma DB.
DB Path (optional)	db_path	Path to the history database. Defaults to an in-memory SQLite database.

Parameters

Attribute	Parameter	Description
Memory Config	memory_config	Memory configuration.

Example Usage

from eclipse.memory import Memory

memory = Memory(memory_config={"llm_client": llm_client})

In this example, a Memory instance is created with a specified Large Language Model client. This setup allows the system to retain and manage context effectively across various agent interactions within the pipeline.