1
0
mirror of https://github.com/huggingface/diffusers.git synced 2026-01-27 17:22:53 +03:00

Fix: Align VAE processing in ControlNet SD3 training with inference (#11909)

Fix: Apply vae_shift_factor in ControlNet SD3 training
This commit is contained in:
Hengyue-Bi
2025-07-15 02:54:38 +08:00
committed by GitHub
parent 3c8b67b371
commit 6398fbc391

View File

@@ -1330,7 +1330,7 @@ def main(args):
# controlnet(s) inference
controlnet_image = batch["conditioning_pixel_values"].to(dtype=weight_dtype)
controlnet_image = vae.encode(controlnet_image).latent_dist.sample()
controlnet_image = controlnet_image * vae.config.scaling_factor
controlnet_image = (controlnet_image - vae.config.shift_factor) * vae.config.scaling_factor
control_block_res_samples = controlnet(
hidden_states=noisy_model_input,