Closed mkmuellner closed 1 year ago
Hey dude, I found the line that they are referencing there. its in SuperAGI\SuperAGI\Jobs\Agent_executor.py ` try: if parsed_config["LTM_DB"] == "Pinecone": memory = VectorFactory.get_vector_storage("PineCone", "super-agent-index1", OpenAiEmbedding()) else: memory = VectorFactory.get_vector_storage("PineCone", "super-agent-index1", OpenAiEmbedding()) except: print("Unable to setup the pincone connection...") memory = None
user_tools = session.query(Tool).filter(Tool.id.in_(parsed_config["tools"])).all()`
Ah thanks! Probably needs an update in the readme.md
Thanks we'll update this in the readme.md!
Thanks we'll update this in the readme.md!
It might be worth checking the format of that part of readme.md while you're there. Currently, it displays as a wall of small text, like the legal disclaimers you see on websites that nobody reads. The actual setup instructions are buried among the developer credits. I'm guessing an HTML tag got deleted.
@TsundokuJim Sure we will have a look, you can also raise a PR for the same we can review the same.
We have solved this issue. Thus closing this.
The readme.md says: "If you're on the Pinecone free plan, you only have 1 pod and 1 index available. As a workaround, change the index name used in test.py where memory is defined: memory = VectorFactory.get_vector_storage("PineCone", "my-current-indexname", OpenAiEmbedding())"
However test.py does not seem to contain a line where memory is defined. There is a line in the agent_config_values where "memory_window":10 is stated, but that seems different from what the above suggests. Any ideas?