| """ |
| Credit: laksjdjf |
| https://github.com/laksjdjf/cgem156-ComfyUI/blob/main/scripts/attention_couple/node.py |
| """ |
|
|
| import torch.nn.functional as F |
| import math |
|
|
|
|
| def repeat_div(value: int, iterations: int) -> int: |
| for _ in range(iterations): |
| value = math.ceil(value / 2) |
|
|
| return value |
|
|
|
|
| def get_mask(mask, batch_size, num_tokens, original_shape): |
| """ |
| Credit: hako-mikan |
| https://github.com/hako-mikan/sd-webui-regional-prompter/blob/main/scripts/attention.py |
| |
| Issue Found/Fixed by. arcusmaximus & www |
| https://github.com/arcusmaximus/sd-forge-couple/tree/draggable-box-ui |
| """ |
|
|
| image_width: int = original_shape[3] |
| image_height: int = original_shape[2] |
|
|
| scale = math.ceil(math.log2(math.sqrt(image_height * image_width / num_tokens))) |
| size = (repeat_div(image_height, scale), repeat_div(image_width, scale)) |
|
|
| num_conds = mask.shape[0] |
| mask_downsample = F.interpolate(mask, size=size, mode="nearest") |
| mask_downsample = mask_downsample.view(num_conds, num_tokens, 1).repeat_interleave( |
| batch_size, dim=0 |
| ) |
|
|
| return mask_downsample |
|
|
|
|
| def lcm(a, b): |
| return a * b // math.gcd(a, b) |
|
|
|
|
| def lcm_for_list(numbers): |
| current_lcm = numbers[0] |
| for number in numbers[1:]: |
| current_lcm = lcm(current_lcm, number) |
| return current_lcm |
|
|
