2022-11-23 (Summary of last few weeks)
-
New storage class; S3 support. We now have a generic storage class, which
allows for special URLs anywhere anywhere you can usually specify a URL,
e.g.CHECKPOINT_URL,dest_url(after dreambooth training), and the new
MODEL_URL(see below). URLs like “s3:///bucket/filename” will work how
you expect, but definitely read docs/storage.md to understand the format better. Note in particular the triple forwardslash (“///”) in the beginning to use the default S3 endpoint. -
Dreambooth training, working but still in development. See this forum post for more info.
-
PRECISIONbuild var, defaults to"fp16", set to""to use the model defaults (generally fp32). -
CHECKPOINT_URLconversion:- Crash / stop build if conversion fails (rather than unclear errors later on)
- Force
cpuloading even for models that would otherwise default to GPU.
This fixes certain models that previously crashed in build stage (where GPU
is not available). -
--extract-emaon conversion since these are the more important weights for
inference. -
CHECKPOINT_CONFIG_URLnow let’s to specify a specific config file for
conversion, to use instead of SD’s defaultv1-inference.yaml.
-
MODEL_URL. If your model is already in diffusers format, but you don’t
host it on HuggingFace, you can now have it downloaded at build time. At
this stage, it should be a.tar.zstfile. This is an alternative to
CHECKPOINT_URLwhich downloads a.ckptfile and converts to diffusers. -
test.py:- New
--bananaarg to run the test on banana. Set environment variables
BANANA_API_KEYandBANANA_MODEL_KEYfirst. - You can now add to and override a test’s default json payload with:
--model-arg prompt="hello"--call-arg MODEL_ID="my-model"
- Support for extra timing data (e.g. dreambooth sends
train
anduploadtimings).
- New
-
Dev: better caching solution. No more unruly
root-cachedirectory. See
CONTRIBUTING.md for more info.