Xianshun Wang

GSL是2.3，我可以找个时间再试一下，但是你也可以试一下Ubuntu，因为vs的C++和标准C有丢丢区别，有可能是这个原因导致的。

> 好的，谢谢 我装个虚拟机试一下。感谢博主。 > […](#) > ------------------ 原始邮件 ------------------ 发件人: "xianshunw/scindapsus" ***@***.***>; 发送时间: 2022年2月23日(星期三) 下午4:47 ***@***.***>; ***@***.******@***.***>; 主题: Re: [xianshunw/scindapsus] please, I use your code in Windows，but My result is very terrible. (Issue #4)...

> 1000? 对不起我实在不记得了，很久之前的事了。

I have the same question, I just want to use 'ad' to generate disparity. I try the command parameters like this `./main.lua kitti ad-a predict -net_fname net/net_kitti_fast_-a_train_all.t7 -left samples/input/kittiL.png -right...

> Hi， In config : text model name is 'pretrained_models/clip-vit-base-patch32-projection'. But I can't load tokenizer or clip_model by model name. I look into huggingface.co, can't find the model. Can you...

Thanks for your reply. The objectness model is the part I'm most interested in.

> > > > Excellent work! Thanks for your contribution! Will the training code be available later? How many computing resources are needed to reproduce your reported results? > >...

After struggling with a bug all morning, I believe I now understand why the shuffle must be set to false here. Let's first take a look at how the engine...

> @xianshunw @avicooper1 Setting `shuffle=True` will not cause anything error in my experiment. I am just curious why it is `False` for default in Pipeline Dataloader Sampler. Apologies, it was...

> I'm not sure why it is also set to False but I agree that `shuffle=True` is better. Making it configurable is better.