About Code release for "SimMTM: A Simple Pre-Training Framework for Masked Time-Series Modeling" (NeurIPS 2023 Spotlight), https://arxiv.org/abs/2302.00861
107
stars
13
forks
source link
Questions about training configuration for classification task #9
I had difficulty reproducing the results from the paper when running run.sh in SimMTM_Classification folder.
Some of the results I obtained are as follows:
SleepEEG -> Epilepsy: acc=94.51
SleepEEG -> FD-B: acc=63.40
Additionally, even when setting the hyperparameters as described in Appendix A.4 of the paper, the results were still the same.
Could you please provide the training configuration that would allow us to reproduce the classification results?
Thank you for sharing wonderful work!
I had difficulty reproducing the results from the paper when running run.sh in SimMTM_Classification folder. Some of the results I obtained are as follows: SleepEEG -> Epilepsy: acc=94.51 SleepEEG -> FD-B: acc=63.40
Additionally, even when setting the hyperparameters as described in Appendix A.4 of the paper, the results were still the same.
Could you please provide the training configuration that would allow us to reproduce the classification results?