add datasets>=2.9.0 requirement

replacing fs for load_from_disk with storage_options

remove deprecated averaged_model

add conditional training

fix when save_model_epochs != save_images_epochs

add missing pathlib import

migrate to diffusers!

temporary patch

remove unnecessary check

get ready for migration to diffusers

add sample_rate and n_fft params

remove unnecessary imports

add attribution

train latent dm with pre-trained vae from hf hub

add max_epochs arg

add max_epochs arg

added training notebook for colab

log after x epochs, not after 1

fix for latest version of diffusers

add sample_rate parameter

fix audio logging for VAE

move resolution specification to dataset generation

Merge branch 'main' of https://github.com/AI-Guru/audio-diffusion into AI-Guru-main

update readme

Enables non square mel spectrograms

get working for monochrome

take channels into account

take channels into account

handle steps correctly

various fixes

fixes

update readme

fix pipelines

tidy

fix progress bar

update train_unconditional for latent diffusion

update train_unconditional for latent diffusion

merge with diffusers latest version

added vae notebook

convert to hf model