search

experiencor / keras-yolo2

Easy training on custom dataset. Various backends (MobileNet and SqueezeNet) supported. A YOLO demo to detect raccoon run entirely in brower is accessible at https://git.io/vF7vI (not on Windows).
MIT License
1.73k stars 784 forks source link

Minor issue. This argv is never used. I guess it should be args? #385

Open ysyyork opened 5 years ago

ysyyork commented 5 years ago

https://github.com/experiencor/keras-yolo2/blob/4e8c85ce02435f136d4f4cfe930b4ccb759fbaf8/gen_anchors.py#L103

Very minor but I guess this should be args? code works because args are global. Not big deal haha

robertlugg commented 5 years ago

I agree. Perhaps even move the parsing to main to avoid the global:

def main():
    args = argparser.parse_args()
    config_path = args.conf
    num_anchors = args.anchors