nn-revisit
nn-revisit copied to clipboard
paper issue
Hi, thanks for your wonderful work! I have a question that whether your proposed KnnFocalLike loss is better than Focal loss. I do not find the comparison experiment in your paper.
Hi, thanks for the question. We did not run that experiment since the goal is to show how to incorporate knn results into current practice. But I do see your point, it will be interesting to compare with focal loss. Will get back to you in a few days after I get the result
Hi, thanks for the question. We did not run that experiment since the goal is to show how to incorporate knn results into current practice. But I do see your point, it will be interesting to compare with focal loss. Will get back to you in a few days after I get the result
Thanks, If KnnFocalLike loss is better than Focal loss, we will use it in our work. I look forward to your experimental results