Error running demo
Opened this issue · 2 comments
swoooxx commented
nqanh commented
You're calling the default caffe version which doesn't have the RoIAlign layer (it may exist in your system if you install caffe before). Make sure to build caffe from our repo and set PYTHONPATH correctly.
swoooxx commented
@nqanh
Hi there, I cannot train using AffordanceNet because my GPU is only ~6GB.
So now, I am trying to reproduce the results stated in one of your research papers 'Detecting Object Affordances with Convolutional Neural Networks' .
What I have got now is this...
The dataset is the same as the one you used in your paper with 7 classes. Why is my per class accuracy stuck at 0 all the time?
I would greatly appreciate if you could share the code or explain what can be done to solve this issue. Thank you.