Closed duanLH closed 7 years ago
Hi, We haven't tried training these models with only VOC datasets But the example prototxts are just samples.
If you want better mAP, I suggest you increase stepsize, lower gamma (e.g. gamma 0.3165 and stepsize 300000) or adopt 'plateau' LR policy (#7).
@sanghoon Oh, thank you very much
I train the model in example_train_384 using default parameter, by the result of test is only 0.719, Is this because of the trained dataset is not COCO+VOC07+VOC12? thanks