use OFA-CN-Large pre-training model to infer Image Captioning

OFA-Sys / OFA

Official repository of OFA (ICML 2022). Paper: OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework

Apache License 2.0

2.39k stars 248 forks source link

use OFA-CN-Large pre-training model to infer Image Captioning #349

Open Duanmu0312 opened 1 year ago

Duanmu0312 commented 1 year ago

Hello, I found that Image Caption based on MUGE seems to be over-fitting. I want to know how to use OFA-CN-Large pre-training model to infer Image Captioning?

JustinLin610 commented 1 year ago

How do you evaluate overfitting? Is it about its performance biased to e-commerce or something else?