MongoDB Memory ¤
For chat that requires a persistent and scalable message history, you can use a MongoDB database to store the conversation history.
Set MongoDB memory¤
import declarai
from declarai.memory import MongoDBMessageHistory
gpt_35 = declarai.openai(model="gpt-3.5-turbo")
@gpt_35.experimental.chat(
chat_history=MongoDBMessageHistory(
connection_string="mongodb://localhost:27017/mydatabase",
session_id="unique_chat_id")
) # (1)!
class SQLBot:
"""
You are a sql assistant. You help with SQL related questions with one-line answers.
"""
sql_bot = SQLBot()
- The
connection_string
parameter specifies the connection details for the MongoDB database. Replacelocalhost
,27017
, andmydatabase
with your specific MongoDB connection details. Thesession_id
parameter uniquely identifies the chat session for which the history is being stored.
Set MongoDB memory at runtime¤
In case you want to set the MongoDB memory at runtime, you can use the set_memory
method.
import declarai
from declarai.memory import MongoDBMessageHistory
gpt_35 = declarai.openai(model="gpt-3.5-turbo")
@gpt_35.experimental.chat
class SQLBot:
"""
You are a sql assistant. You help with SQL related questions with one-line answers.
"""
sql_bot = SQLBot(chat_history=MongoDBMessageHistory(connection_string="mongodb://localhost:27017/mydatabase",
session_id="unique_chat_id"))
Dependencies¤
Make sure to install the following dependencies before using MongoDB memory.
pip install declarai[mongodb]