EfficientPose Training

[nishata24]

Hi, I am a student and I want to use EfficientPose. I have been working on the 6th step of the installation process. Below is what I get after running the command: python setup.py build_ext --inplace.

(EfficientPose) user@Puget-225743:~/EfficientPose$ python setup.py build_ext --inplace
running build_ext

Also, when training phi = 0 EfficientPose model on object 8 of Linemod (driller) using COCO pretrained weights, I get nan values as shown in the screenshot below.

Do you know what might be going on and how I can solve this issue? Thank you.

[enesdoruk]

i experinced same thing. Firstly i used nvidia a100 graphic card and results is nan however when i use gtx1080, results is not nan and logical

[Strknvdie39]

Hi, I am a student and I want to use EfficientPose. I have been working on the 6th step of the installation process. Below is what I get after running the command: python setup.py build_ext --inplace.

(EfficientPose) user@Puget-225743:~/EfficientPose$ python setup.py build_ext --inplace running build_ext

Also, when training phi = 0 EfficientPose model on object 8 of Linemod (driller) using COCO pretrained weights, I get nan values as shown in the screenshot below.

Do you know what might be going on and how I can solve this issue? Thank you.

I am a studentt oo
And can you tell me how to compile Cython modules using gcc on Ubuntu??? I got an error as image below
Thank you,

[julingers]

Hi, I am a student and I want to use EfficientPose. I have been working on the 6th step of the installation process. Below is what I get after running the command: python setup.py build_ext --inplace.

(EfficientPose) user@Puget-225743:~/EfficientPose$ python setup.py build_ext --inplace running build_ext

Also, when training phi = 0 EfficientPose model on object 8 of Linemod (driller) using COCO pretrained weights, I get nan values as shown in the screenshot below.

Do you know what might be going on and how I can solve this issue? Thank you.

Did you solve the problem? I meet too and I used 3060 laptop GPU. It didn't work.

[tiny-ych-wolf]

Because 3060 can just support CUDA>11, however, the needed cuda version is 10.0.
So you may need to change your GPU.