Open PeterLuoCoder opened 1 year ago
When I train the DANet model, the results are shown in the figure, and the results do not reach the results in Table 10 of your paper.
When I train the DANet model, the results are shown in the figure, and the results do not reach the results in Table 10 of your paper.
BatchSize is a very important param that affects the accuracy. You may consider setting it according to the advice in the paper
Hi, your test code is a multi-GPU test, but I'm using a single GPU for training, how can I change the test code? I ask you to help me.
Set the testing shell like this, CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node=2
I used:
CUDA_VISIBLE_DEVICES=0 python -m torch.distributed.run --nproc_per_node=1 --master_port 29506 test.py --dataset vaihingen --val_batchsize 16 --models swinT --head mlphead --crop_size 512 512 --save_dir work_dir --base_dir ../../ --information num1
but got an error:
test.py FAILED
Can you help me?
Hi, your test code is a multi-GPU test, but I'm using a single GPU for training, how can I change the test code? I ask you to help me.