gluon-cv icon indicating copy to clipboard operation
gluon-cv copied to clipboard

Published 20 hours ago • verified publisher icon dmlc

Add dim check while calculating mIoU&pixAcc for segmentation test

Open chunhanl opened this issue 5 years ago • 4 comments

  1. The test.py example mentioned at segmentation.rst does not match the arguments mentioned in test.py should probably be change to python test.py --dataset ade20k --pretrained --model fcn --backbone resnet50 --eval

  2. The dimension doesn't seem right when calculating the pixACC and mIoU The output is 3D and the target is 2D The safest way I could think of is to add a new axis if batch dim is missing. Might have a better way to deal with.

Result after modification (without training) Screenshot from 2019-10-17 20-26-19

chunhanl avatar Oct 18 '19 04:10 chunhanl

Apologies for the new spaces, it was automatically modified by Pycharm IDE

chunhanl avatar Oct 18 '19 05:10 chunhanl

Job PR-991-1 is done. Docs are uploaded to http://gluon-vision-staging.s3-website-us-west-2.amazonaws.com/PR-991/1/index.html Code coverage of this PR: pr.svg vs. Master: master.svg

mli avatar Oct 18 '19 05:10 mli

python test.py --dataset ade20k --pretrained --model fcn --backbone resnet50 --eval Screenshot from 2019-10-18 16-35-31

python test.py --dataset pascal_voc --pretrained --model fcn --backbone resnet101 --eval Screenshot from 2019-10-18 16-31-43

chunhanl avatar Oct 18 '19 23:10 chunhanl

Job PR-991-2 is done. Docs are uploaded to http://gluon-vision-staging.s3-website-us-west-2.amazonaws.com/PR-991/2/index.html Code coverage of this PR: pr.svg vs. Master: master.svg

mli avatar Oct 18 '19 23:10 mli