The code is running super slow?

[Haiyan-Chris-Wang]

@WangYueFt @syb7573330 I could run the code successfully, but the code is running super slow. The speed is about 10 epochs/day. Do you have any idea about this problem or it is the normal speed for this code?

[WangYueFt]

Are you running on CPU? What's the GPU model if it's runing on GPU?

[Haiyan-Chris-Wang]

@WangYueFt Yeah, I found the problem. Now I have switched to the GPU (GTX 1080Ti), although It is faster than CPU, the speed is still slow, about 5 epochs/hour. So how fast is your running speed?

[WangYueFt]

I think that's roughly the same.

[Haiyan-Chris-Wang]

I see. Thanks for your reply!

[Haiyan-Chris-Wang]

@WangYueFt @syb7573330 Could you help me explain why do we need to perform tf.concat here to combine the previous net? It seems didn't mention in the paper.

dgcnn/models/dgcnn.py

Line 79 in 29948ad

net = tf_util.conv2d(tf.concat([net1, net2, net3, net4], axis=-1), 1024, [1, 1],

[KangchengLiu]

@Haiyan-Chris-Wang How did you switch to GPU？