search

dangweili / pedestrian-attribute-recognition-pytorch

A simple baseline for pedestrian attribute recognition in surveillance scenarios
334 stars 81 forks source link

strange error #13

Closed 2014gaokao closed 5 years ago

2014gaokao commented 5 years ago

after I run sh script/experiment/train.sh

Traceback (most recent call last): File "./script/experiment/train_deepmar_resnet50.py", line 18, in from baseline.dataset import add_transforms ImportError: No module named baseline.dataset

baseline.dataset exists, it puzzles me

dangweili commented 5 years ago

maybe you can put this code in the start of train_deepmar_resnet50.py sys.path.append(os.getcwd())

2014gaokao commented 5 years ago

Thank you! I use the second version of RAP dataset and get result as follows Label-based evaluation: mA: 0.7498 Instance-based evaluation: Acc: 0.7099, Prec: 0.8252, Rec: 0.8114, F1: 0.8182 I have no idea if it is a good result and does this research area mature now? I wonder if i can do some improvments such as using gcn to model label depedencies etc.