search

hizhangp / yolo_tensorflow

Tensorflow implementation of YOLO, including training and test phase.
MIT License
799 stars 442 forks source link

Can not detect object after training your network ? #87

Closed ngthanhtin closed 5 years ago

ngthanhtin commented 5 years ago

Hello, I have modified your net to train only 2 objects which are dog and person, but after training process, I get the loss is around 6 and then I tested (inference) the pre-trained model, the net can not detect anything. Do you have encounter this problem ? If yes, please tell me where is wrong. Anyway, thanks for your contribution.

Benzzzxxx commented 5 years ago

Maybe your train data is not enough or train epochs is not enough.

LinzzMichael commented 5 years ago

I have meet the same problem,but my loss was very high,it's about 10,how about yours?

Unhapppy commented 5 years ago

i met the same problem, have you solved it?

Jeremy-lf commented 5 years ago

so am i ,what's your solution?

ngthanhtin commented 5 years ago

i dont know, either

eefresher commented 5 years ago

i also have same problem ,my loss is acound 7

mxazz123 commented 5 years ago

The loss is too high. Just uncomments the WEIGHTS in yolo/config file, and retrain it

ngthanhtin commented 5 years ago

thank you very much :D