Comments (1)
Additionally, how does nnU-Net handle the absence of a particular label class during training and validation?
It is not compulsory to have each label in each case. This is for training/val in general.
Furthermore, the mean dice score on my test set was 0.827 (calculated by averaging the four categories and then averaging the four average values). I also conducted an external test on 116 cases, where the mean dice score was 0.83. This seems strange to me,
Why would this be strange?
About the calculation, simply leave out the nans for those cases. you may average X cases for a certain label and Y cases for another.
from nnunet.
Related Issues (20)
- How to Choose the Best Model Checkpoints and Folds for nnUNetPredictor? HOT 3
- RuntimeError: Could not find a task with the ID HOT 4
- Trying to train with GPU instead of CPU HOT 12
- Fatal FIPS Failure HOT 1
- Training with patience HOT 1
- dimension of predictions results from nnunet differs from the labels
- Confusion about Cascade Training HOT 3
- Command not found
- ConvTranspose 3D is not supported on MPS HOT 10
- Did not find the expected number of training cases however I have 10 training cases (snapshots of error and json attached) HOT 1
- The program gets stuck in the CPU during prediction and cannot produce a result. HOT 1
- Not sure where this warning is coming from or what it means? HOT 2
- Issue with moving plans from target dataset and transfering it to the pretraining dataset: HOT 2
- Data naming convention
- Title: KeyError: '1'
- What should the dataset.json be like for a rgb 2d images? HOT 1
- nnUNet configuration. HOT 2
- Final Contouring HOT 3
- My error in train 3dfullres HOT 2
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 nnunet.