r""" Instantiate a [`BrushNetModel`] from [`UNet2DConditionModel`]. Parameters: unet (`UNet2DConditionModel`):

r""" Instantiate a [`BrushNetModel`] from [`UNet2DConditionModel`]. Parameters: unet (`UNet2DConditionModel`):

Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/

Get the order of each step for sampling by the singlestep DPM-Solver.

Constructs an exponential noise schedule.

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

Constructs an polynomial in log sigma noise schedule.

Constructs a continuous VP noise schedule.

Gets the current EMA decay rate.

Apply passed in transforms for HuggingFace Datasets.

Numpy array to tensor. Args: imgs (list[ndarray] | ndarray): Input images. bgr2rgb (bool): Whether to change bgr to rgb.