How long does it take for the model to converge after training?

teticio / audio-diffusion

Apply diffusion models using the new Hugging Face diffusers package to synthesize music instead of images.

GNU General Public License v3.0

707 stars 69 forks source link

How long does it take for the model to converge after training? #5

Closed 980202006 closed 2 years ago

980202006 commented 2 years ago

I want to train this model on my own dataset, how long does it take for the model to converge?

teticio commented 2 years ago

Hi. You can see the tensorboard here https://huggingface.co/teticio/audio-diffusion-instrumental-hiphop-256/tensorboard, for example, and see the loss curves and generated audios along the way. It already generates decent results after 10 epochs, which takes about 12 hours on my RTX 2080 Ti with this dataset of around 30,000 samples.