Comments (5)
zhijiejia
commented on September 17, 2024
Oh, I have a idea that may I make the pre training parameters of vgg16 are initialized again. .....
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.
zhijiejia
commented on September 17, 2024
Another problem I found is that when the above parameter initialization method is used, the network can't be non convergent. I don't know why。。
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.
zhijiejia
commented on September 17, 2024
It may be that dropout, p = 0.6, brings too much instability. p value should be reduced appropriately.
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.
GeoZcx
commented on September 17, 2024
We use the pre-trained VGG16 parameter as the deep feature extraction network parameter, only the feature fusion network needs initialization. If initialize the whole network as you do, the network would take a long time to converge.
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.
fpan98
commented on September 17, 2024
how to only initialize parameter except vgg16
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.
Related Issues (20)
- question about DSIFN HOT 1
- question about batchsize HOT 1
- question about Input channel HOT 1
- question about your data set HOT 1
- Problem about the network needs to be independently trained in two steps HOT 1
- Q About Result weight files HOT 1
- train val test file HOT 3
- Can you share the complete code? HOT 6
- 为什么下载下来的数据集，train和val部分的mask全黑呢？ HOT 2
- About the train.py and predict.py HOT 6
- 请问怎么获得完整代码 HOT 5
- 数据集mask全黑 HOT 5
- Could the raw big images be made publiced？ HOT 1
- Google drive or Onedrive version of dataset HOT 2
- The label of the CDD dataset is not binary map. HOT 1
- model HOT 4
- Regarding training code
- not found train code HOT 1
- how can i use t和、 HOT 1
- 预训练模型跑出的predicted image和论文里给出的效果差很多
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from a-deeply-supervised-image-fusion-network-for-change-detection-in-remote-sensing-images.