Guidance on Using run_experiment vs. run_evals in Phoenix

it reads there is quite some similarity between run_experiment and run_evals, is there any general guidance on when to use which?

from phoenix.experiments import run_experiment
from phoenix.evals import run_evals

· Sorted by Oldest

it reads there is quite some similarity between run_experiment and run_evals, is there any general guidance on when to use which?

from phoenix.experiments import run_experiment
from phoenix.evals import run_evals

· Sorted by Oldest

Zhuyi X.
·
run_experiments is based on Dataset, while run_evals is based on pandas dataframe.
Xander S.
·
Hey Zhuyi X. 👋 It depends on what you are trying to accomplish. run_experiments is our newer API that will allow you to run experiments over datasets and see those scores in the UI. run_evals will allow you to annotate traces, spans, and documents.
💡1

Zhuyi X.
·
run_experiments is based on Dataset, while run_evals is based on pandas dataframe.
Xander S.
·
Hey Zhuyi X. 👋 It depends on what you are trying to accomplish. run_experiments is our newer API that will allow you to run experiments over datasets and see those scores in the UI. run_evals will allow you to annotate traces, spans, and documents.
💡1