q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens

`scatter` along one dimension and `gather` along another.

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Get keypoint predictions from score maps. Note: batch_size: N num_keypoints: K heatmap height: H heatmap width: W

Sets the discrete timesteps used for the diffusion chain (to be run before inference). Args: num_inference_steps (`int`):

Zero out the parameters of a module and return it.

Initialize the RMSNorm normalization layer. Args: dim (int): The dimension of the input tensor. eps (float,

Returns the smallest integer greater than or equal to 'number' that is divisible by 'factor'.

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Merge the video and audio into a new video, with the duration set to the shorter of the two, and overwrite the original video file. Para

Returns the closest integer to 'number' that is divisible by 'factor'.

Rescales the image so that the following conditions are met: 1. Both dimensions (height and width) are divisible by 'factor'. 2. The to

Calculate the normalized distances between preds and target. Note: batch_size: N num_keypoints: K dimension of keypoints:

Draw a rounded rectangle with transparency.

Returns the largest integer less than or equal to 'number' that is divisible by 'factor'.

r initialize sequence parallel group.

Get final keypoint predictions from heatmaps and apply scaling and translation to map them back to the image. Note: num_keypoints: K

r""" Reconstruct video tensors from patch embeddings. Args: x (List[Tensor]): List of patchified features

Configures a model object. This includes setting evaluation modes, applying distributed parallel strategy, and handling device placem

Modulate heatmap distribution with Gaussian. sigma = 0.3*((kernel_size-1)*0.5-1)+0.8 sigma~=3 if k=17 sigma=2 if k=11; sigma~=1.5

Initialize a per-GPU cross-attention cache.

Initialize a Per-GPU KV cache for the SelfAttn.

"Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the prediction of x_0 at t

"Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the prediction of x_0 at t

Get center and scale of bounding box from bounding box. The expected format is [min_x, min_y, max_x, max_y].

Crop image according to the supplied bounding box. res: [rows, cols]

Performs distributed attention based on DeepSpeed Ulysses attention mechanism. please refer to https://arxiv.org/pdf/2309.14509 Args:

Draw keypoints and connections representing hand pose on a given canvas. Args: canvas (np.ndarray): A 3D numpy array representing th

Extract WASD actions from translation in c2w poses. Args: c2ws: Array of shape (N, 4, 4) threshold: Minimum translation

r""" Generates video frames from input image and text prompt using diffusion process. Args: input_prompt (`str`):