Open Tangzy7 opened 7 years ago
How did you calculate mAP? On current solution I have 53.4%. But maybe my implementation of mAP is wrong.
@Firyuza Hi, I used the same script with Faster RCNN paper. https://github.com/rbgirshick/py-faster-rcnn/blob/master/lib/datasets/voc_eval.py
@Tangzy7 Hi, thanks. I try it.
Do you know do we have to count FaslePositive if score of bounding box is less then score threshold?
@Firyuza I think the official script of calculating mAP already did it.
if ovmax > ovthresh:
if not R['difficult'][jmax]:
if not R['det'][jmax]:
tp[d] = 1.
R['det'][jmax] = 1
else:
fp[d] = 1.
else:
fp[d] = 1.
@Tangzy7 Yes, I know that they do it, but I'm not sure is it right?
@Tangzy7 Did you finally retrain on pascal voc? and did you achieve the result claimed in yolo9000 paper?
However, set R['det'][jmax] = 1 will not change the value in class_recs[image_ids[d]]. Hence, for the next same image, doing R = class_recs[image_ids[d]] will give a new R which has R['det'][jmax] = 0. Is this understanding correct?
I retrain the last layer using VOC2007+2012 trainval dataset. It worked fine and I got the mAP = 0.74. Personally, I recommend not to use the retrain.py, I think there are some bugs in it.
@Tangzy7 @KLH1472 ,Hi, I want to run the retrain_yolo.py, but I don't know how to convert the VOC data into .npz format, can you help me?
I wanted to achieve the benchmark of the yolov2 paper, which is a 78% mAP on voc2007+voc2012. I used retrain_yolo.py to train on voc2007+2012 and test on voc2007_test, but I only got a mAP of 37%. Did anyone recurrent the result successfully?