FengMu1995

@chdzhen 你好,我也是和你一样在android手机部署模型,不知道是手机旧了还是其他什么原因,在用opencl加载的时候出现 SVM capalibilties: SVM_FINE_GRAIN_BUFFER SVM capalibilties: SVM_ATOMICS SVM capalibilties: SVM_FINE_GRAIN_BUFFER SVM capalibilties: SVM_ATOMICS SVM capalibilties: SVM_FINE_GRAIN_BUFFER SVM capalibilties: SVM_ATOMICS SVM capalibilties: SVM_FINE_GRAIN_BUFFER SVM capalibilties: SVM_ATOMICS Segmentation fault 你知道是啥原因吗

Hi, I want to know how to label the dataset? What tools do you use to label the dataset?

@Smeegol 问下用tnn比用coreml快么

> I've got a better performance on my own dataset,whats U should do is being patient and down your lr to 0.01,0.005 and 0.001 问下你用的什么损失函数在预训练的基础上再训练的

Yes, the multiple datasets are combined into one big dataset which be used to train or the datasets are trained one by one.

@ToBigboss 你测过cpu占比高么,若是可以达到实时但是cpu占比高可能也是用不了

Hi, I wonder the rewriten code is right? It is because I've met the same question.

> I'm trying to train Real-ESRGAN with 400 images. The training curves are very noisy, especially the perceptual loss (l_g_precep). I've tried to use various different learning rates. I'm using...

try to add tl.set_backend('pytorch') into your code

理解了,谢谢大佬