Lei-Lei Ma
Lei-Lei Ma
You can view it as a coefficient
voc2012 ? http://host.robots.ox.ac.uk:8080/
> > > Your work is awesome. > Could you please release code to reproduce? It should be released soon
in train_autodeeplab.py replace parser.add_argument('--gpu-ids', nargs='*', type=int, default=0, help='which GPU to train on (default: 0)') with parser.add_argument('--gpu_ids', nargs='*', type=str, default='0', help='which GPU to train on (default: 0)') --gpus_ids type=str default='0'
CUDA_VISIBLE_DEVICES=0,1,2 python train_autodeeplab.py --backbone resnet --lr 0.007 --workers 2 --epochs 40 --batch_size 1 --gpu_ids 0,1,2 --eval_interval 1 --base_size 64 --crop_size 64 --dataset cityscapes RuntimeError: CUDA out of memory. Tried to...
> > > thanks for your feedback, for further investigation, could you please execute `log_v=1 python -m jittor.test.test_core` 
LD_PRELOAD=/usr/lib/libstdc++.so.6 python -m jittor.test.test_example `https://pastebin.ubuntu.com/p/79jdS98927/`
I gave up when I couldn't downgrade