SAR_plane

predictions_dict = model.predict(tf.expand_dims(resized_image_tensor, 0)) recognitions = model.postprocess(predictions_dict)

1

代码中可以找到predict（）和postprocess（）函数的具体实现吗？我不太理解 @abstractmethod def predict(self, preprocessed_inputs, scope=None): pass @abstractmethod def loss(self, predictions_dict, scope=None): pass @abstractmethod def postprocess(self, predictions_dict, scope=None): pass 这几个函数这么定义了之后要怎么用？

The loss is so large and iou of each class is 0 except background.

5

I train segnet using SBD and pascal voc 2012, but my loss is so large such as 578002. And when i eval my model on val, i get iou of...

config.yml for segnet

2

I want to train segnet, but i don't know how to set the config. Can anyone help me? Thank you so much. What's more, how should i process the dataset...

Under what coordinate system is the normal corresponding to each plane in the dataset marked? In my follow-up experiment, I need to use the predicted normal. How to determine its...

Which file can I find the network structure? I want to know how to design FCIS on MXNET. Thank you!

Segmentation fault (core dumped)

5

I use Pytorch1.0.0,cuda9.0,gcc4.9.2,my driver version is 396.26 so I can't use cuda 10. When I run `demo.py`,I get this.  Should I update my gcc again? or are there any...