WongKinYiu / yolor

implementation of paper - You Only Learn One Representation: Unified Network for Multiple Tasks (https://arxiv.org/abs/2105.04206)
GNU General Public License v3.0
1.98k stars 524 forks source link

What does box loss mean during training? #259

Open udnqv opened 2 years ago

udnqv commented 2 years ago

Hi, I'm training a model. During the training there are three values displayed: box, obj and cls. Could someone explain to me what exactly these three describe? On the last two values I can imagine what I means, but i have absolutely no clue what 'box' means. Help would be highly appreciated. Thank you in advance!