Comments (4)
Thank you!
from enjoy-hamburger.
Hi! Thank you for being so interested in the Hamburger! I guess that you need to import van
in your train/test.py
file to register the VAN backbone. For example, you may use the /seg_light_ham/dist_test.sh
instead of the script under the /tools/
directory because this command applies to the correct train/test.py
portal.
BTW, you can deploy the Light-Ham (VAN-Base) model. This one balances efficiency and performance very well.
Please feel free to ping me if you have additional questions!
from enjoy-hamburger.
Thank you!
One more question: The batch size I inputed should be one, but when I was checking the data, there is actually 12 images batched together. Do you know where is the setting for that?
By the way, do you have model pretrained on cityscapes?
from enjoy-hamburger.
Hi! I am not sure about the test batch size. The test batch size may be related to the number of GPUs you use because it is actually DistributedDataParallel (bs 1 per GPU to enable multi-GPU MSFlip test). Perhaps you can try to use one GPU only.
Good question. I only have pretrained checkpoints on ADE20K because some computing got expired. The good news here is that one of my collaborators will release a new paper regarding the power of VAN+Ham on segmentation. You may expect this paper.
Thanks!
from enjoy-hamburger.
Related Issues (11)
- Applying Hamburger to other models makes the training collapse soon HOT 2
- Confusion on update rules of Ham HOT 1
- About Fix point iteration HOT 3
- question about the implementation of one-step gradient
- ZeroDivisionError: float division by zero
- Arxiv version or blog HOT 5
- BatchNormalization HOT 1
- 缺少主干网络 HOT 2
- Default process group has not been initialized HOT 4
- Difference between the code and Eq(13) in the paper about the gradient calculation HOT 5
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 enjoy-hamburger.