Bagheera

snapshotting top-level "bpool" filesystem causes grub to fail

33

### System information Type | Version/Name --- | --- Distribution Name | Gentoo Distribution Version | amd64 Kernel Version | 5.19.6-gentoo Architecture | amd64 OpenZFS Version | zfs-kmod-2.1.99-1358_g60d995727 / zfs-2.1.99-1359_gede037cda...

DDIM produces incorrect samples with SDXL (epsilon or v-prediction)

37

### Describe the bug When generating images with SDXL and DDIM, there is some residual noise in the outputs. This leads to a "smudgy" look, and in cases where fewer...

Mixture-of-Experts partial diffusion implementation for base SD 1.x / 2.x pipelines

32

# What does this PR do? This pull request ports our denoising_start code to the text2img pipeline, and the denoising_start and denoising_end code from the img2img pipeline. This brings legacy...

Feature request: ability to truncate datasets on split rather than pad

2

As a Developer working on latent diffusion model training via SimpleTuner, it has become evident that the built-in mechanism for splitting datasets across processes is not smart enough to apply...

add mps and xpu to examples

2

also moved it to bf16 which results in a substantial speedup over fp32

[show and tell] apple mps support

7

with newer pytorch (2.4 nightly) we get bfloat16 support in MPS. i tested this: ```py from parler_tts import ParlerTTSForConditionalGeneration from transformers import AutoTokenizer import soundfile as sf import torch device...

At least for SD2.1, freezing the first _n_ layers (typically, 17) allows it to learn more effectively without catastrophic loss / destructive moves to the base layers. This helps the...

[mps] training / inferencing deepfloyd must be done in float32

3

### Describe the bug _**Just to keep track of this issue, because I'm not sure if I've done something wrong or if it's due to the current issues in Pytorch's...

[mps] training / inference dtype issues

32

when training on Diffusers without attention slicing, we see: ``` /AppleInternal/Library/BuildRoots/ce725a5f-c761-11ee-a4ec-b6ef2fd8d87b/Library/Caches/com.apple.xbs/Sources/MetalPerformanceShaders/MPSCore/Types/MPSNDArray.mm:788: failed assertion `[MPSNDArray initWithDevice:descriptor:] Error: total bytes of NDArray > 2**32' ``` but with attention slicing, this error disappears....

apple: support for MLX quantized linear in diffusers

1

**Is your feature request related to a problem? Please describe.** As an Apple MPS user, it always feels somewhat like we're second-class citizens with respect to the latest and greatest...