For some tests we also need the underlying UNet. For these, we'll build the UNetControlNetXSModel from the UNet and ControlNetXS-Adapter

Makes broadcastable attention and causal masks so that future and masked tokens are ignored. Arguments: attention_mask (

Get the model modules inside the transformers library.

Get the objects in module that are models.

r""" The function to the pipeline for image inversion as described by the [LEDITS++ Paper](https://arxiv.org/abs/2301.12247). If the s

r""" Load pretrained attention processor layers into [`UNet2DConditionModel`]. Attention processor layers have to be defined in

This will load the LoRA layers specified in `state_dict` into `text_encoder` Parameters: state_dict (`dict`):

Convert a numpy image or a batch of images to a PIL image.

Concatenate x and its positional encodings, following NeRF. Reference: https://arxiv.org/pdf/2210.04628.pdf

r""" Recursively replace all instances of `LoraLayer` with corresponding new layers in `model`.

Updates paths inside attentions to the new naming scheme (local renaming)

Save a model and its configuration file to a directory, so that it can be re-loaded using the `[`~models.adapter.MultiAdapter.from_pr

Sets a specific adapter by forcing the model to only use that adapter and disables the other adapters. If you are not familiar with

r""" Sets the attention processor to use to compute attention. Parameters: processor (`dict` of `AttentionProcessor` or o

r""" Sets the attention processor to use to compute attention. Parameters: processor (`dict` of `AttentionProcessor` or o

Sets the noise scales used for the diffusion chain (to be run before inference). The sigmas control the weight of the `drift` and `di

r""" Set the switch for the npu flash attention.

Sort a list of objects following the rules of isort (all uppercased first, camel-cased second and lower-cased last). Args: objec

Disables the fused QKV projection if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip>

Implementation closely follows k-diffusion. https://github.com/crowsonkb/k-diffusion/blob/6ab5146d4a5ef63901326489f31f1d8e7dd36b48/k_diffusio

Implementation closely follows k-diffusion. https://github.com/crowsonkb/k-diffusion/blob/6ab5146d4a5ef63901326489f31f1d8e7dd36b48/k_diffusio

Constructs the noise schedule of Karras et al. (2022).

Constructs the noise schedule of Karras et al. (2022).

Constructs the noise schedule of Karras et al. (2022).

Fused `Conv2d()` followed by `downsample_2d()`. Padding is performed only once at the beginning, not between the operations. The fused op is c