↓ 1 callersMethodget_fold_unfold :param x: img of size (bs, c, h, w) :return: n img crops of size (n, bs, c, kernel_size[0], kernel_size[1])
imaginairy/modules/diffusion/ddpm.py:965
↓ 1 callersFunctionget_gaussian_kernel2d(
kernel_size_x: int, kernel_size_y: int, sigma_x: float, sigma_y: float, dtype: DType, device: Device
imaginairy/vendored/refiners/fluxion/utils.py:80
↓ 1 callersFunctioninterpolate_images(
image_list,
scale=1.0,
fps_multiplier=2,
model_weights_path=None,
device=None,
)
imaginairy/enhancers/video_interpolation/rife/interpolate.py:363
↓ 1 callersFunctionintersect We resize both tensors to [A,B,2] without new malloc: [A,2] -> [A,1,2] -> [A,B,2] [B,2] -> [1,B,2] -> [A,B,2] Then we compute the area of
imaginairy/vendored/facexlib/detection/retinaface_utils.py:79
↓ 1 callersFunctionlatent_to_raw_image Converts a tensor of size (1, 4, x, y) into a PIL image of size (x*4, y*4). Args: tensor (numpy.ndarray): A tensor of size (1, 4, x, y).
imaginairy/utils/log_utils.py:509