Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

I do not get the expected results #16

Open
mikeletxeberria opened this issue Jan 26, 2021 · 1 comment
Open

I do not get the expected results #16

mikeletxeberria opened this issue Jan 26, 2021 · 1 comment

Comments

@mikeletxeberria
Copy link

Hi, when I try to execute the code as it is, I can not achieve your results. The system and configuration I am using are the following ones:

OS: Windows 10
Conda env:

backcall=0.2.0=pypi_0
certifi=2020.6.20=py36_0
cudatoolkit=10.0.130=h5327add_6
cudnn=7.6.5=cuda10.0_0
cupy=6.0.0=py36h230ac6f_0
cupy-cuda102=9.0.0a2=pypi_0
cycler=0.10.0=pypi_0
easydict=1.9=pypi_0
fastrlock=0.5=pypi_0
intel-openmp=2020.3=h57928b3_311
joblib=0.17.0=pypi_0
kiwisolver=1.3.1=pypi_0
libblas=3.8.0=21_mkl
libcblas=3.8.0=21_mkl
liblapack=3.8.0=21_mkl
matplotlib=3.0.3=pypi_0
mkl=2020.4=hb70f87d_311
numpy=1.16.2=pypi_0
opencv-python=4.4.0.46=pypi_0
pillow=8.0.1=pypi_0
pip=20.2.4=py36haa95532_0
pyparsing=2.4.7=pypi_0
python=3.6.12=h5500b2f_2
python-dateutil=2.8.1=pypi_0
python_abi=3.6=1_cp36m
pyyaml=5.3.1=pypi_0
scikit-learn=0.23.2=pypi_0
scipy=1.5.4=pypi_0
setuptools=50.3.1=py36haa95532_1
six=1.15.0=pyh9f0ad1d_0
sklearn=0.0=pypi_0
sqlite=3.33.0=h2a8f88b_0
threadpoolctl=2.1.0=pypi_0
torch=1.4.0=pypi_0
torchvision=0.5.0=pypi_0
tqdm=4.54.0=pypi_0
vc=14.1=h0510ff6_4
vs2015_runtime=14.16.27012=hf0eaf9b_3
wheel=0.35.1=pyhd3eb1b0_0
wincertstore=0.2=py36h7fe50ca_0
zlib=1.2.11=h62dcd97_4

Command: python run.py -d options\kitti\kitti_default_configuration.yml

I have just changed the references to the KITTI dataset folder and the align_corners=True. The result is the following (even the ground truth is not printed as shown in the paper) :

map

Thank your for your help,

Mikel Etxeberria

@Huangying-Zhan
Copy link
Owner

Hi Mikel, could you check if the flow model is named correctly?
The flow model should be named as "network-default.pytorch" in the current implementation because there is a hard-coded setup which recognises the name.

# FIXME: hardcode for network-default.pytorch

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants