The results of the model in the MMMO dataset failed to reproduce.
JackAILab opened this issue · comments
Dear authors:
I can't get the results mentioned in your paper, is there something wrong with the parameters of my model? I did not change the results of the model and the original data.
Image One: Your Results:
Image Two: My Results:
Image One: Acc Evaluate:
Image Three: Parameter setting:
Generally, the results will fluctuate. But 40% accuracy on test set is too low and unnormal. I think this issue may be caused by some other reasons, maybe you can tune the learning rate, gamma, batch size or layers.