audio-diffusion / scripts

Commit History

  1. add datasets>=2.9.0 requirement

    507b852

  1. replacing fs for load_from_disk with storage_options⁠

    Steve
    Unverified signature
    960050d

  1. remove deprecated averaged_model

    7bd9ee1

  1. add conditional training

    f29faf1

  1. fix when save_model_epochs != save_images_epochs

    fdb8eec

  1. add missing pathlib import

    eef3fb2
  2. migrate to diffusers!

    8a3fb2e

  1. temporary patch

    52ca0bd

  1. remove unnecessary check

    29c19d2

  1. get ready for migration to diffusers

    f30235e

  1. add sample_rate and n_fft params

    5bc60f9

  1. remove unnecessary imports

    e0ef439
  2. add attribution

    f8a240c
  3. train latent dm with pre-trained vae from hf hub

    af8111a

  1. add max_epochs arg

    b3e97c5
  2. add max_epochs arg

    a8d2b62

  1. added training notebook for colab

    8bf6ce9
  2. log after x epochs, not after 1

    1bb3e27
  3. fix for latest version of diffusers

    07ba0fd

  1. add sample_rate parameter

    d7aaca6

  1. fix audio logging for VAE

    c5dcd04

  1. move resolution specification to dataset generation

    327bccf
  2. Merge branch 'main' of https://github.com/AI-Guru/audio-diffusion into AI-Guru-main

    e3b5a6d

  1. update readme

    01c4a98
  2. Enables non square mel spectrograms

    1b5fc25
  3. get working for monochrome

    529c646

  1. take channels into account

    4f552a8
  2. take channels into account

    13aa297
  3. handle steps correctly

    1ef9d1c

  1. various fixes

    2561128

  1. fixes

    21c77d0

  1. update readme

    d122744
  2. fix pipelines

    96e542f
  3. tidy

    001a426

  1. fix progress bar

    399a445
  2. update train_unconditional for latent diffusion

    f15cb42
  3. update train_unconditional for latent diffusion

    8f292f9
  4. merge with diffusers latest version

    9a9737e
  5. added vae notebook

    7aaaf62
  6. convert to hf model

    8aa7c27