Value Error

[utimcraig]

Hi, I'm trying to run the nanobody notebook -code block 12- and I'm encountering this error:

`Rounds (+ Final Round): 67%|████████████████████████████████████████▋ | 4/6 [02:17<01:12, 36.08s/it]
Team: 0%| | 0/2 [00:08<?, ?it/s]
Rounds (+ Final Round): 67%|████████████████████████████████████████▋ | 4/6 [02:25<01:12, 36.40s/it]

---

ValueError Traceback (most recent call last)
Cell In[12], line 10
3 print(f"Number of summaries: {len(project_specification_summaries)}")
5 project_specification_merge_prompt = create_merge_prompt(
6 agenda=project_specification_agenda,
7 agenda_questions=project_specification_questions,
8 )
---> 10 run_meeting(
11 meeting_type="individual",
12 team_member=PRINCIPAL_INVESTIGATOR,
13 summaries=project_specification_summaries,
14 agenda=project_specification_merge_prompt,
15 save_dir=project_specification_dir,
16 save_name="merged",
17 temperature=CONSISTENT_TEMPERATURE,
18 model=model,
19 num_rounds=num_rounds,
20 )
21 print("completed")

File ~/virtual_lab/src/virtual_lab/run_meeting.py:231, in run_meeting(meeting_type, agenda, save_dir, save_name, team_lead, team_members, team_member, agenda_questions, agenda_rules, summaries, contexts, num_rounds, temperature, model, pubmed_search, return_summary)
229 # Check run status
230 if run.status != "completed":
--> 231 raise ValueError(f"Run failed: {run.status}")
233 # If final round, only team lead or team member responds
234 if round_index == num_rounds:

ValueError: Run failed: failed`

It's unclear what is wrong here- perhaps the meeting status is not being changed correctly in the project specification section?

Edit: it seems like sometimes the selection discussions don't get to 100% but code moves on anyway

[swansonk14]

Hi @utimcraig,

Thank you for raising this issue!

The only time I've seen that run failed issue is when my OpenAI API key has run out of funds and required a refill. Is there any chance you're having the same issue?

If that's not the case, then you may need to do a little debugging of the run object to figure out what's going on since this seems to be a failure related to the OpenAI API rather than the Virtual Lab. This page has some information on the different OpenAI run status possibilities. The "failed" status says "You can view the reason for the failure by looking at the last_error object in the Run. The timestamp for the failure will be recorded under failed_at." Can you see what that last_error object says?

As a side-note, as you likely saw in the Jupyter notebook, many of the discussion are run in parallel using concurrent.futures, so you may want to change that and run just one discussion at a time while debugging.

Thanks,
Kyle