autogen.agentchat.contrib.capabilities.teachability.MemoStore

MemoStore

MemoStore(
    verbosity: int | None = 0,
    reset: bool | None = False,
    path_to_db_dir: str | None = './tmp/teachable_agent_db'
)

Provides memory storage and retrieval for a teachable agent, using a vector database.
Each DB entry (called a memo) is a pair of strings: an input text and an output text.
The input text might be a question, or a task to perform.
The output text might be an answer to the question, or advice on how to perform the task.
Vector embeddings are currently supplied by Chroma’s default Sentence Transformers.

Parameters:

Name	Description
verbosity	Type: int | None Default: 0
reset	Type: bool | None Default: False
path_to_db_dir	Type: str | None Default: ’./tmp/teachable_agent_db’

Instance Methods

add_input_output_pair

add_input_output_pair(
    self,
    input_text: str,
    output_text: str
) -> 

Adds an input-output pair to the vector DB.

Parameters:

Name	Description
input_text	Type: str
output_text	Type: str

get_nearest_memo

get_nearest_memo(self, query_text: str) -> 

Retrieves the nearest memo to the given query text.

Parameters:

Name	Description
query_text	Type: str

get_related_memos

get_related_memos(
    self,
    query_text: str,
    n_results: int,
    threshold: int | float
) -> 

Retrieves memos that are related to the given query text within the specified distance threshold.

Parameters:

Name	Description
query_text	Type: str
n_results	Type: int
threshold	Type: int | float

list_memos

list_memos(self) -> 

Prints the contents of MemoStore.

prepopulate

prepopulate(self) -> 

Adds a few arbitrary examples to the vector DB, just to make retrieval less trivial.

reset_db

reset_db(self) -> 

Forces immediate deletion of the DB’s contents, in memory and on disk.