Pedro Cuenca
Pedro Cuenca
Thanks for this terrific repository! I'm confused about the configuration of patches when using the `blindsr` dataset type. The H and L patch sizes are extracted from the configuration here:...
- [x] I have searched to see if a similar issue already exists. **Is your feature request related to a problem? Please describe.** As a workaround for the issue described...
- [x] I have searched to see if a similar issue already exists. Possibly related to #1379. **Is your feature request related to a problem? Please describe.** The use case...
This is mostly a question. I find it a little weird that the output of `encode` in `AutoencoderKL` is the diagonal Gaussian distribution from which the hidden states are obtained,...
I'm working on this for my own use, so I can train in Google Cloud TPU instances. So far I've done the minimum adjustments necessary to make it work for...
Upload is controlled by a new option `logger`/`wandb`/`log_model`. When enabled, model files are uploaded to W&B as artifacts attached to the current run, at the frequency specified by `save_checkpoint_freq`. I'm...
This is useful for users that belong to several W&B teams.
Like in the Imagen Video paper. According to Katherine Crowson it's more stable and faster: https://twitter.com/rivershavewings/status/1578193039423852544?s=21&t=oIToMoHbtU1U8VH3q4Mzng
Case in point: attempting to load a pipeline that uses the k-LMS scheduler using `from_pretrained` without `scipy` installed fails [here](https://github.com/huggingface/diffusers/blob/v0.4.1/src/diffusers/pipeline_utils.py#L457) because `load_method_name` is `None`. The error the user sees is:...
The Stable Diffusion pipeline accepts latents as an optional input for reproducibility purposes, see #247. In conversation with @tmabraham, he suggested extending this feature to other pipelines too. In addition...