runzeer

when using the DataParallel, the memory cost in each GPU is not equal. So if convenient, could you change the mode to DDP?

Could you provide an example for the config file? ------------------ 原始邮件 ------------------ 发件人: "Tian Zhi"; 发送时间: 2019年7月26日(星期五) 中午11:19 收件人: "tianzhi0549/FCOS"; 抄送: "张润泽";"Mention"; 主题: Re: [tianzhi0549/FCOS] Hi ， excellent work！ In...

I mean if I change the RPN with the FCOS in the two-stage framework, it works? So in your paper, you just did the experiments with the only RPN rather...

> I do the related experiments , I make FCOS as RPN in faster-rcnn ,I only get mAP 34~ when res50 as backbone, It may not work as well as...

I mean the training hyperparameters such like training steps, the training tokens per batch. In the CogView paper, it gave us the relevant parameters. 

Thanks a lot! Also, I had a question. when the model can generate the visible image, the loss could be what level. I mean could you give me some training...

thx for your excellent work! Now I have a tensorflow framework.So can I use the crfrnnlayer as a layer behind the layers from tensorflow such as slim?

Also, I want to know the training cost when training on the larger dataset like LAION. Could you tell me the training epochs when training the larger models and the...

@zaynmi Have you solved the problem? I encountered the same problem. If you solved it, could you please tell me the reason?

@airsplay I checked the testing log. It has nothing with the results. And the evaluation is finished but the accuracy is 0. So if convenient, could you help find the...