search

AlexeyAB / yolo2_light

Light version of convolutional neural network Yolo v3 & v2 for objects detection with a minimum of dependencies (INT8-inference, BIT1-XNOR-inference)
MIT License
303 stars 116 forks source link

BIT1-XNOR accuracy #13

Open ramanathan831 opened 6 years ago

ramanathan831 commented 6 years ago

How far is the accuracy when BIT1-XNOR is used for training and inference

ghimiredhikura commented 6 years ago

Hello @ram1897, I am also trying to test BIT1-XNOR. So far I got 25.39% accuracy in voc test set using TinyYolov2. Here is cfg file that I used for training and testing.
tiny-yolo-obj_xnor_new.zip result
image

I would also like to know @AlexeyAB comment on this result.

Thanks.

Best, Deepak

attiamohammed commented 4 years ago

Thanks for sharing your results. I keep trying to do the training. However, the prediction MAP is always 0%. The network is unable to detect any object. Are there any suggestions?