Given a prompt and a list of kwarg schedules, return a series of prompts that follow the schedule. kwarg_schedules example: { "p

Replace a value in a nested dictionary using a path.

Resample fine and coarse segmentation tensors to the given bounding box and derive labels for each pixel of the bounding box Args:

Resamples U and V coordinate estimates for the given bounding box Args: u (tensor [1, C, H, W] of float): U coordinates v (t

Constructs a ResNet-50 model_hyper.

For compatibility with schedulers that need to scale the input according to the current timestep.

select the proper image to show for each frame of a video.

For creating seamless tile images. Args: tile_mode: One of "", "x", "y", "xy". If "x", the image will be tiled horizonta

Shorten the tree representation to a given number of lines around a given line index.

Encodes the image in slices (for lower memory usage).

Adjusts the bounding box to make it as close to a square as possible while ensuring it does not exceed the max_size of the image and still in

Updates the step count.

Convert a Tensor to a PIL Image. The tensor must have shape `[1, channels, height, width]` where the number of channels is either 1 (gra

Function: ---------- apply the inverse of affine transform 'trans' to uv Parameters: ---------- @trans: 3x3 np.array

Process an image by tiling it, processing each tile, and then merging them back into one image. Args: img (Tensor): The input image tens