r""" Transformer block used in Hunyuan-DiT model (https://github.com/Tencent/HunyuanDiT). Allow skip connection and QKNorm Parameters:

r""" Processor for implementing scaled dot-product attention (enabled by default if you're using PyTorch 2.0). This is used in the TripoSG mod

r""" Processor for implementing scaled dot-product attention (enabled by default if you're using PyTorch 2.0) with fused projection layers. Th

r""" Processor for implementing scaled dot-product attention (enabled by default if you're using PyTorch 2.0). This is used in the MIDI model.

A logging filter to add color to certain log levels.

The sin/cosine positional embedding. Given an input tensor `x` of shape [n_batch, ..., c_dim], it converts each feature dimension of `x[..., i]` i

Output class for the scheduler's `step` function output. Args: prev_sample (`torch.FloatTensor` of shape `(batch_size, num_channels,

r""" Output class for TripoSG pipelines.

Type for instantiating a function with keyword arguments

Pipeline for image-to-scene generation based on pre-trained shape diffusion.

The rectified flow scheduler is a scheduler that is used to propagate the diffusion process in the rectified flow. This model inherits from

r""" Helper for DiffusionPipeline with vae and transformer.(mainly for DIT)

TripoSG: Diffusion model with a Transformer backbone. Inherit ModelMixin and ConfigMixin to be compatible with the sampler StableDiffusionPi