Results 6 comments of varghese

@Cartucho Any updates on this, I am seeing a 20 % difference.

@erikguo can you please let me know how you installed this repo. I git cloned this repo, 1) make clean 2) make 3) sh build.sh [ 96%] Building CUDA object...

I get lower score upon `./build/tools/caffe train --solver=models/release/pelee_coco_voc/solver.prototxt --weights=models/release/pelee_coco_voc/pelee_304x304_acc7637.caffemodel` I0322 03:14:51.507519 267 caffe.cpp:155] Finetuning from models/pelee_coco_voc/pelee_304x304_acc7637.caffemodel I0322 03:14:51.560392 267 upgrade_proto.cpp:77] Attempting to upgrade batch norm layers using deprecated params: models/pelee_coco_voc/pelee_304x304_acc7637.caffemodel...

any thoughts?

Hi; I was able to get the score @Robert-JunWang mentions. There was something wrong with the LMBD I created.

How do I pass 38x38 feature map instead of the duplicate 19x19 feature map?