search

yeungchenwa / FontDiffuser

[AAAI2024] FontDiffuser: One-Shot Font Generation via Denoising Diffusion with Multi-Scale Content Aggregation and Style Contrastive Learning
https://yeungchenwa.github.io/fontdiffuser-homepage/
299 stars 26 forks source link

about training time #28

Closed ChenSiyi1 closed 9 months ago

ChenSiyi1 commented 11 months ago

How long does it take to train this model from scratch?

yeungchenwa commented 10 months ago

Hi @ChenSiyi1, about 4 days under the capacity of the training data in paper, and it is conducted on a single 3090.