search

aliyun / conditional-lane-detection

Apache License 2.0
396 stars 89 forks source link

train --no-validate #31

Open hzzzzjzyq opened 3 years ago

hzzzzjzyq commented 3 years ago

eventhough i try --validate with one gpu ,still returned non-zero exit status 1.

conditional-lane-detection$ CUDA_VISIBLE_DEVICES=0 PORT=29001 /mnt/data/pycharmmm/conditional-lane-detection/tools/dist_train.sh /mnt/data/pycharmmm/conditional-lane-detection/configs/condlanenet/culane/culane_small_train.py 1 --validate

huoxingdawang commented 3 years ago

I had a similar problem, I got the error "TypeError:'DataContainer' object is not subscriptable", and then it stopped.

huoxingdawang commented 2 years ago

@hzzzzjzyq I saw you closed the issue 19 hours ago. I want to know if you have solved the issue? How is it solved?

shimmerjordan commented 2 years ago

After trying several versions of pytorch, I found that torch>=1.8 may cause such issue. It works when I downgrade torch to 1.7.0. Noted that my cuda version is 10.2