Pytorch-Deeplab

Pytorch-Deeplab copied to clipboard

The normalization in data pre-processing

2

I notice that the normalization of images in your code uses "image - mean," rather than the normal "(image - mean)/std". Is there a specific reason for using this kind...

yzou2

CuDNN error: CUDNN_STATUS_INTERNAL_ERROR

5

hi, i use it to train for my own data set, but it run error, can u give me some suggestion? (my segmentation class img is gray image). /opt/conda/conda-bld/pytorch_1532581333611/work/aten/src/THCUNN/SpatialClassNLLCriterion.cu:99: void...

J-zin

XUYUNYUN666

I download this weight and share on google drive https://drive.google.com/open?id=19_PGrzARnJugSKULbTyRJ0txlI2UPvm6. 讲真，装百度网盘的时候真的超级怕被在系统里开个后门。

MaureenZOU

Slightly lower validation accuracy in Pytorch 1.0.0

6

Hi @speedinghzl , I used your pytorch deeplab v2 implementation, same setting, with pytorch version of 1.0.0. The validation mIoU of VOC_scenes_20000.pth is 71.1. Is this degradation due to randomness?...

yzou2

Hi, i am sorry but where can i find the crf implementation?

joanacmferreira

multiscale model has lower IOU performance

1

Hi @speedinghzl , I used your pytorch deeplab v2 implementation, same setting, with torch version of 0.4.0, I used batch size of 16 to train the multi-scale model and evaluated...

ymli39

Frozen the statistics of BN layers?

5

Hello, Do you implement "frozen BN layers"? I find that the running means and running vars still keep changing during training.

nihaomiao

CUDA error: out of memory

2

Hi, By not changing anything on the code, i ve ran train.py for the augmented pascal voc dataset. But im getting CUDA error. Might you have a suggestion on how...

siinem

https://github.com/speedinghzl/Pytorch-Deeplab/blob/89a49c2f340cf359cfadab29c899a74ade53ee70/train.py#L129-L135 I think this part of code is wrong since it will yeild same parameters multiple times.

wheatdog