Ciao Luca B.! You can do this by configuring the following the environment variable PHOENIX_COLLECTOR_ENDPOINT to point to the server running in a different process or container. https://docs.arize.com/phoenix/environments
will try this out and come back if I have questions - thanks in the meantime!
Hey there! I am not using OpenAI to work with phoenix, is there a way for me to write my own instrumentor? I already have a wrapper for llamaindex. https://docs.arize.com/phoenix/environments#container
Hey Luca! You certainly could! You are looking to trace things that wrap LlamaIndex? What are you looking to instrument? If so you can certainly publish your own spans via the tracer - We are definitely going to support more custom instrumentation but let us know if you hit any walls! https://github.com/Arize-ai/phoenix/blob/main/src/phoenix/trace/tracer.py
Thanks again for the reply! I found what I needed by inspecting the source code in llamaindex:
from phoenix.trace.exporter import HttpExporter
import llama_index
os.environ["PHOENIX_COLLECTOR_ENDPOINT"] = "http://0.0.0.0:6006"
exporter = HttpExporter(endpoint="http://0.0.0.0:6006")
llama_index.set_global_handler("arize_phoenix", exporter=exporter)
This should configure phoenix as global handler and export the traces to the UI running in a different process (e.g.a container)
Awesome! Glad you figured that out :)
Would you like to add this to your documentation? I could help if you wanted.
Will ping you know later this week (bit busy with PyData Global tonight).
