gek
> How many images do you use to compute FID? This line becomes slower with the number of images you use, but unless you use a really large number of...
> Realized there was a Kubernetes config issue on my end, Im able to run mixtral properly now using 2 X A100 80GB GPUs, sorry about that! Could you please...
@JMGama Me too, are you fix this ?
I also encountered this problem. The name looks like
There is a long gap between the validation accuracy of the dataset of vlmevalkit and the model paper
I've noticed that most of the results are dissimilar compared to those in the research paper. I believe that the framework that's in use should be rectified, making it more...
这个指标是越高越好，还是越低越好？或者说，多个model，在相同的数据集上进行imageReward的评测，是不是得分越高越好？
> Hi, can you report your torchmetrics version? > > ``` > pip list | grep torchmetrics > ``` torchmetrics 1.3.1
Here is my metric result. second column is avg_score
> Just realized the code implementation in dev branch (where we run the experiment) is different. Now it is aligned with our paper. Thanks for catching this bug! > >...
The Clipscore indicator is still a little different. Very tiny