Closed zijing-wu closed 5 years ago
The version at this commit Fixes #20 seems to work (with the provided god_save_the_queen.mid file).
But I can't get as good results as in the demos using the "--ext big" models explained in the examples. It is even worse if I use the pretrained model. Maybe there's some parameter tuning work to do, or there are further essential improvements after that commit.
Hi, I just pushed a new PyTorch version. I hope that this solves your problems. The architecture has not been optimized at all but it still gives decent results.
Hi, First, thank you for the great work! I followed the first a few examples to train the models and got some outputs with default raw dataset. Now, I would like to input my own .mid file but face the error:
I noticed #43 also encountered the same problem but there seems not to have new commits after the issue. I am wondering if any updates will be available soon? Also, I found that #6 also tried to supplied customer melody but was able to get valid outputs. May I know if some older versions of the project work? Would you suggest using an old version to get it work with -m option for now?
Thank you!