How to configure the progress bar for updates?

[Ben81828]

Hello, thanks for your excellent work!

I'm currently running VQA_RAD and PATH-VQA. Despite metrics have finished calculated, the progress bar fails to update.

Could you please advise on how to configure it for updates?

[corentin-ryr]

Hello,

Thank you for your feedback.

Unfortunately, I am unable to reproduce the problem. I ran this minimal code and got the following output:

from multimedeval import EvalParams, MultiMedEval, SetupParams
import json
import logging
logging.basicConfig(level=logging.INFO)

def batcher(prompts):
    return ["dummy answer" for _ in range(len(prompts))]

engine = MultiMedEval()

config = json.load(open("config.json"))
engine.setup(setupParams=SetupParams(**config))

results = engine.eval(["VQA-Rad"], batcher, evalParams=EvalParams(batch_size=4, fewshot=True))

INFO:MultiMedEval:***** Benchmarking : VQA-Rad *****
INFO:MultiMedEval:Running inference:   0%|                                                                                                                                                                    | 0/113 [00:00<?, ?it/s]
INFO:MultiMedEval:Running inference: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 113/113 [00:00<00:00, 153.00it/s]
INFO:MultiMedEval:{'bleu': 0.0, 'F1': 0.0, 'recall': 0.0, 'closedQuestionsAccuracy': 0.0, 'openQuestionsRecall': 0.0, 'openQuestionsAccuracy': 0.0}

Could you provide a minimal example to reproduce your problem?