Dalao

@Shimingyi 我刚下载pretrain模型，效果也很差。可以说脚步几乎没有看出任何优化的痕迹，很滑很飘，而且跳动的时候几乎看不出在跳。与paper视频效果差异极大。可能是什么原因？

@Duankaiwen 有三个任务同时训的结果了吗？

@Duankaiwen 老铁

@yanzhuangzhuang-beep No

@zhiqwang I think I forget to initPlugin when parsing trt since it contains BatchedNMSPlugin. However, I actually want set the model input width height when convert, rather than export. So...

Still not avaiable?

I just run the example, it not work on iOS...... Unhandled Exception: PlatformException(bgTaskSchedulingFailed(Error Domain=BGTaskSchedulerErrorDomain Code=1 "(null)") error, Scheduling the task using BGTaskScheduler has failed.

Hi, am ask does the current training codes only for conditioned iamge geneation rather than text iamge generation? Does the tokenizer training code release as well?

same request

Yes. I made it. it works OK. it works both on mobile and PC. add my telegram to request my fork of this feature. lucifertrump