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
How can I train my own data set in an inmage description task? #439
I want to train myself on an image description task with a dataset that includes both images and text to get the weights for this task