PKU-YuanGroup / Video-LLaVA

Video-LLaVA: Learning United Visual Representation by Alignment Before Projection
https://arxiv.org/pdf/2311.10122.pdf
Apache License 2.0
2.86k stars 207 forks source link

video eval error #114

Closed wsxwd closed 7 months ago

wsxwd commented 7 months ago

while runing CUDA_VISIBLE_DEVICES=0 bash scripts/v1_5/eval/run_qa_msrvtt.sh raise an error

ValueError: temperature has to be a strictly positive float, but is 0.0

should we set do_sample=False or set temperature > 0?

LinB203 commented 7 months ago

Yes.

wsxwd commented 7 months ago

Yes.

set do_sample=False, right?

LinB203 commented 7 months ago

Yes.

set do_sample=False, right?

Sorry for reply above. Set do_sample=False and temperature=0.0.

wsxwd commented 7 months ago

Great thanks