MTCNN-Tensorflow
MTCNN-Tensorflow copied to clipboard
Why you use "replace=True" in gen_imglist_pnet.py ?
There is a statement for choose negative examples “neg_keep = npr.choice(len(neg), size=len(neg), replace=True)” in gen_imglist_pnet.py. The augment "replace=True" will lead to repetitive examples, so why not use "replace=False" to get more effective examples?
Agree with @creatist , same doubt on this line.
agree