diffusers/pr_12652/en/api/models/easyanimate_transformer3d.md

EasyAnimateTransformer3DModel

A Diffusion Transformer model for 3D data from EasyAnimate was introduced by Alibaba PAI.

The model can be loaded with the following code snippet.

from diffusers import EasyAnimateTransformer3DModel

transformer = EasyAnimateTransformer3DModel.from_pretrained("alibaba-pai/EasyAnimateV5.1-12b-zh", subfolder="transformer", torch_dtype=torch.float16).to("cuda")

EasyAnimateTransformer3DModel[[diffusers.EasyAnimateTransformer3DModel]]

diffusers.EasyAnimateTransformer3DModel[[diffusers.EasyAnimateTransformer3DModel]]

Source

A Transformer model for video-like data in EasyAnimate.

Parameters:

num_attention_heads (int, defaults to 48) : The number of heads to use for multi-head attention.

attention_head_dim (int, defaults to 64) : The number of channels in each head.

in_channels (int, defaults to 16) : The number of channels in the input.

out_channels (int, optional, defaults to 16) : The number of channels in the output.

patch_size (int, defaults to 2) : The size of the patches to use in the patch embedding layer.

sample_width (int, defaults to 90) : The width of the input latents.

sample_height (int, defaults to 60) : The height of the input latents.

activation_fn (str, defaults to "gelu-approximate") : Activation function to use in feed-forward.

timestep_activation_fn (str, defaults to "silu") : Activation function to use when generating the timestep embeddings.

num_layers (int, defaults to 30) : The number of layers of Transformer blocks to use.

mmdit_layers (int, defaults to 1000) : The number of layers of Multi Modal Transformer blocks to use.

dropout (float, defaults to 0.0) : The dropout probability to use.

time_embed_dim (int, defaults to 512) : Output dimension of timestep embeddings.

text_embed_dim (int, defaults to 4096) : Input dimension of text embeddings from the text encoder.

norm_eps (float, defaults to 1e-5) : The epsilon value to use in normalization layers.

norm_elementwise_affine (bool, defaults to True) : Whether to use elementwise affine in normalization layers.

flip_sin_to_cos (bool, defaults to True) : Whether to flip the sin to cos in the time embedding.

time_position_encoding_type (str, defaults to 3d_rope) : Type of time position encoding.

after_norm (bool, defaults to False) : Flag to apply normalization after.

resize_inpaint_mask_directly (bool, defaults to True) : Flag to resize inpaint mask directly.

enable_text_attention_mask (bool, defaults to True) : Flag to enable text attention mask.

add_noise_in_inpaint_model (bool, defaults to False) : Flag to add noise in inpaint model.

Transformer2DModelOutput[[diffusers.models.modeling_outputs.Transformer2DModelOutput]]

diffusers.models.modeling_outputs.Transformer2DModelOutput[[diffusers.models.modeling_outputs.Transformer2DModelOutput]]

Source

The output of Transformer2DModel.

Parameters:

sample (torch.Tensor of shape (batch_size, num_channels, height, width) or (batch size, num_vector_embeds - 1, num_latent_pixels) if Transformer2DModel is discrete) : The hidden states output conditioned on the encoder_hidden_states input. If discrete, returns probability distributions for the unnoised latent pixels.