guojinma

I tried to convert movenet's tflite model to TNN format, but there exist some unsupported operators in the tflite model, such as Less, Greater, And, Not and so on. As...

wav2lip+face_enhancement效果很差

1

我尝试使用wav2lip_hq来驱动人脸图片的时候，如果加入face_enhancement参数，效果就会变的如下图的奇怪效果，参数用的是wav2lip文档中的默认参数，而不加face_enhancement的时候并没有这种问题。请问是否face_enhancement的处理逻辑有些地方没有更新？

using multi gpu

1

Is there anyone made it work on multi gpu? In my case, it just consumes the memory.

can I run it on a server without pygame and deskptop?

3

I'm new to opengl or pygame. The server is a centos6 which has no desktop installed, and failed to install pygame with many difficulties. When i comment out the pygame...

AdaCos的问题

3

你的实现中theta_med是计算batch内所有样本在所有类别上的角度均值，论文中说是“the median of all corresponding classes’ angles”，我理解的是类似每个样本在标签类上的夹角，不知道对不对？

questions about shakedrop

1

In the original paper, shakedrop is given as **G(x) = x + (bl + a + bl*a)F(x)**, but your application makes **out = (drop_factor + alpha + alpha * (1...

a more radical resize rate?

1

Have you ever tried a more radical resize rate, such as 76x76 or 56x56?

Will the blur kernels be released?

5

I wanna genenarte some blured face images with the kernels and train with that, will you release the kernels?

slow on 3090 and very high cpu usage

6

I have a 3090 GPU, and converted the falcon-40b-instruct and quantized by Q3_K. But when I run the test, prediction is 3x slower than the reported, then I check the...