Kumar Manas

icon company Free University of Berlin icon location Berlin icon location Working for traffic rule knowledge representation and explainable knowledge for autonomous driving.

Results 6 issues of Kumar Manas

BNN taking longer time then full precision Network

6 comment

I was trying to compare Larq BNN and full precision ( by making Integer and kernel_quantizer= None). I found that time taken to run the program is more for BNN...

question

BNN taking longer time than Full Precision

While running your BNN code It takes 1.5 times more time compare to the Full precision Keras model for same architecture. Is it excepted or I am missing something?

efficient net as feature extractor

Thanks for your torch implementation. I am interested in getting features of image same as we do in Resnet. Is it possible to do via your implementation? If yes, can...

Are training and evaluations both Quantized?

In the implementation, it looks like training is Quantized form but evaluation/test is still in Full precision. Is this assumption correct?

Net implemented in BNN

Did you implement any particular type of net-like VGG or Alexnet or some other custom network for this code?

scene graph encoding choice

From paper, it seems that scene graph is in form of text triplet and you encode the text triplet using Graph encoder. Is my assumption true or image features is...