diffengine.models.editors.wuerstchen.efficient_net_encoder

Module Contents

Classes

EfficientNetEncoder

EfficientNet encoder for text-to-image generation.
class diffengine.models.editors.wuerstchen.efficient_net_encoder.EfficientNetEncoder(c_latent=16, c_cond=1280, effnet='efficientnet_v2_s')[source]

Bases: diffusers.models.modeling_utils.ModelMixin, diffusers.configuration_utils.ConfigMixin

EfficientNet encoder for text-to-image generation.

Copied from https://github.com/huggingface/diffusers/blob/main/examples/ wuerstchen/text_to_image/modeling_efficient_net_encoder.py

Parameters:

  • c_latent (int) –

  • c_cond (int) –

  • effnet (str) –

forward(x)[source]

Forward pass.

Parameters:

x (torch.Tensor) –

Return type:

torch.Tensor