-
I don't have access to enough computation resource so I'm training on 8gpus with resolution 1024 on cityscapes, and I find it really hard to recover the result that's even nearly compatible to the one…
-
I am trying to use wider_resnet trunk by setting the following in scripts/train_cityscapes.yml:
snapshot: "ASSETS_PATH/seg_weights/wider_resnet38.pth.tar",
arch: wider_resnet.wider_resnet38_a2,
…
-
Hello, I use resnest50-deeplabv3 model and train from scratch on Cityscapes dataset. During training, the loss value doesn't decrease after reaching 0.1( I use a single titan gpu and batch size of 10…
-
# voc dataset
[375,500]
## theoretical optimal config
min_crop_size=[384,480]
max_crop_size=[384,480]
midnet_scale=10 (input_size=[480,480])
# cityscapes dataset
[1024,2048]
## theoretic…
-
hi,
when executing
`python train.py --evaluate --snapshot checkpoints/best_cityscapes_checkpoint.pth`
I get this error:
`subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-z…
ghost updated
5 years ago
-
DeepLabV3 with cityscapes takes long time to execute and we need to figure out the reason and then add it.
-
Pix2Pix模型用“AtoB”这种格式判断方向：
https://github.com/PaddlePaddle/PaddleGAN/blob/21ee373c2a9481231f50b32e0753d3f543a86396/ppgan/models/pix2pix_model.py#L81
但配置文件里direction的格式却是“a2b”“b2a”：
https://github.co…
-
Thanks your code ! I am new to semi-supervised semantic segmentation. It looks like this is first paper to do this task. So is the evaluation carried **out only on validation dataset**? (PascalVOC a…
lxtGH updated
4 years ago
-
I am getting "Cannot allocate memory error" after around 13-15k iterations while trying to reproduce results on "foggy cityscapes" dataset. I running this code on 4 GPUs with 360G memory.
I can re…
-
Is IMG_MEAN taken from the cityscapes dataset?
IMG_MEAN = np.array((103.939, 116.779, 123.68), dtype=np.float32)
Presumably if I train on my own dataset I would have to precompute the mean, is t…