Inference Speed

[ulucsahin]

I trained a model with provided Wojtek data, inference speed is around 1.4 seconds per frame.

I also trained a model using different data which I generated following the steps mentioned. However, it takes 7 seconds to infer 1 frame. I used same amount of num_rays. What might be causing this slowness?

[sunwoo76]

Hello,

Did you fix the code below?

def query_func(pts): with torch.no_grad(): with torch.cuda.amp.autocast(enabled=self.fp16): sigma = self.model.density(pts.to(self.device))['sigma'] return sigma

In line which outputs the sigma value, the condition which is the input of the self.model.desity() is skipped.
So, it makes the error in my code. How did you fix it?

File "/home/cvlab09/projects/swkim/INSTA-pytorch/insta/utils.py", line 600, in query_func sigma = self.model.density(pts.to(self.device))['sigma'] TypeError: density() missing 1 required positional argument: 'cond'