Model Training #1
Unanswered
NuraddinQJ
asked this question in
Q&A
Replies: 1 comment
-
@NuraddinGit Hello. I am very sorry for the delayed response. I missed your issue in my feed... You can ignore the displayed number of steps and go by desired loss to train your model. The steps counter in this implementation is set to len(train_data) so if you want it to display the correct number of steps per epoch (depending on your settings) you will need to divide it accordingly. Training does not take much (especially on Solo Piano datasets) because they are small usually. So I recommend to check out my improved version of DeBussy with all issues corrected here: https://github.com/asigalov61/DeBussy-X Hope this helps. Alex |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi there,
I'm trying to train the model to to generate MIDI music in the style of Chopin but it takes so much time near 4 days with 350,000 step!! what I'm doing wrong?
My question is could you share more information about the number and length of MIDI files and the size was used in the training and how much time it takes?
Beta Was this translation helpful? Give feedback.
All reactions