Closed caiopetruccirosa closed 1 month ago
From what I understand, the experiment config provided in the repository only corresponds to the ResNet50 model pre-trained on ImageNet with a Cross-Entropy loss head.
Yes.
@paplhjak, could you please provide the experiment config files for all methods and weight initialization strategies (IMDB-WIKI, ImageNet, and Random) mentioned in the paper?
I will look into it.
Thank you very much! I am looking forward to it.
I have some more pressing things to work on this week, but if all goes well, you can expect the files to get added next week.
The configuration files will be added by the end of the week.
The files can now be found in facebase/configs
Thank you very much!
Hello! First of all, thank you very much for the paper.
I am trying to reproduce all the results reported in Table 6 of the paper for my research. However, after reviewing the YAML config files for the experiments, I found that there is only one base experiment provided for each benchmark. This does not cover all nine experiments listed in the sub-rows of Table 6 (Cross-Entropy, Regression, OR-CNN, DLDL, DLDL-v2, SORD, Mean-Var., Unimodal, FaRL + MLP).
From what I understand, the experiment config provided in the repository only corresponds to the ResNet50 model pre-trained on ImageNet with a Cross-Entropy loss head.
@paplhjak, could you please provide the experiment config files for all methods and weight initialization strategies (
IMDB-WIKI
,ImageNet
, andRandom
) mentioned in the paper?Thanks in advance!