GitLab

add cross_attention_dim as an argument

* Extented the ability of ddpm scheduler
to utilize model that also predict the variance.

* Update src/diffusers/schedulers/scheduling_ddpm.py
Co-authored-by: Anton Lozhkov <aglozhkov@gmail.com>

* [Vae and AutoencoderKL clean]

* save intermediate finished work

* more progress

* more progress

* finish modeling code

* save intermediate

* finish

* Correct tests

* Add torch_device to the VE pipeline

* Mark the training test with slow

* Expose schedulers

* Update __init__.py
Co-authored-by: Anton Lozhkov <anton@huggingface.co>

final fixes before release

* Typos

* Credits and roadmap

* Second version

* work in progress, fixing tests for numpy and make deterministic

* make tests pass via pytorch

* make pytorch == numpy test cleaner

* change default tensor format pndm --> pt

up

* up

* change model name

* renaming

* more changes

* up

* up

* up

* save checkpoint

* finish api / naming

* finish config renaming

* rename all weights

* finish really

* organize PNDM tests, begin API change

* clean timestep API PNDM

* update pipeline PNDM

* fix typo

* API clean round 2

* small nit

* Add a decorator for register_to_config

* All models and test

* big purge

* more fixes

* finish for now

* added kwargs for easier intialisation of random model

* initial commit for conversion script

* current debug script

* update

* Update

* done

* add updated debug conversion script

* style

* clean conversion script

* clean ddpm api to match ddim

* correct ve sde class

* update pipeline API for ve sde

* make style

* Apply suggestions from code review
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>