CoinCheung/DeepLab-v3-plus-cityscapes
mIOU=80.02 on cityscapes. My implementation of deeplabv3+ (also know as 'Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation' based on the dataset of cityscapes).
PythonMIT
Issues
- 0
Number of epochs
#20 opened by MaugrimEP - 0
Pretrained model
#22 opened by idinardo - 0
visualize evaluation output
#21 opened by At-Walid - 28
train loss
#5 opened by seeyouagain111 - 0
about train.py
#19 opened by Chihanxx - 8
About BatchNormalization
#3 opened by RainPot - 1
How to label my own image
#14 opened by huidemie - 2
PermissionError: [Errno 13] Permission denied: '/tmp/torch_extensions/inplace_abn'
#11 opened by syp2ysy - 0
- 1
different results with same pth
#12 opened by ywangeq - 0
- 1
evluate.py
#16 opened by csBob123 - 0
Big batchsize cannot improve the performance.
#15 opened by LetsGoFir - 2
ConvBNReLU
#13 opened by LiQiufu - 1
How can i change the batch-size?
#8 opened by SecondLi - 2
errors when running the code
#7 opened by pgu-nd - 1
pretrained model
#4 opened by zzzz94 - 1
Upload trained model
#6 opened by phongnhhn92 - 0
- 0
- 2
How to train the model on single GPU?
#2 opened by zhixuanli - 1
how to test on cityscape
#1 opened by neineimeow