Comments (5)
- could this change wait until Jenny comes back on March 17th, it seems like a big refactor.
- I am sure mask_dir doesn't include the float weight maps, mask_dir is always binary target, in preprocessing mask_dir is binary masks which is used to compute and combine the float weight maps created into the input_dir by setting them in a channel of the input data which is given as data_dir in the config
- given that mask_dir is always binary masks, I don't see why we would need any change
Please correct me if I am wrong
from microdl.
Yes mask_dir is always binary masks, but its role is different in segmentation and regression tasks. Here is my understanding of the role of mask_dir for segmentation and regression:
For segmentation,
mask_dir: binary input supplied by the user; used as the target for training
weight_dir: weight maps generated by mask generator using mask_dir as input; used for computing weighted loss
For regression,
mask_dir: generated by mask generator (more common)m or supplied by the user; used for computing weighted loss (masked loss); not the target for training
weight_dir: N/A
To me the change of the role of mask_dir in different use cases is confusing. We can discuss this with Jenny after she is back. I'll leave it as it is for now for this PR
from microdl.
mask_dir is always the same for segmentation or regression i.e it is optional if the user provides it, weight_dir for segmentation is not a mandatory parameter
from microdl.
I've renamed this issue from "mixed usage of mask_dir" since masks from now on will be stored along with other channels in a zarr store.
However the issue still remains that user specified masks (generated outside of microDL) as input to preprocessing would be a good feature to have.
Prior to implementing this, specifications need to be discussed regarding what format of masks to support.
from microdl.
This is still relevant, but straightforward to do by updating the masks in the ome-zarr store. Closing. cc: @Soorya19Pradeep.
from microdl.
Related Issues (20)
- Config issues on gunpowder dataloading branch HOT 12
- Flatfielding computation error on gunpowder dataloading branch HOT 2
- Improve masking of fluorescence data in preprocessing HOT 2
- configs should use channel names and not channel index HOT 2
- Metadata Structure HOT 2
- z-score HOT 1
- Inference shouldn't use gunpowder HOT 5
- intensity.csv HOT 1
- Convert existing tensorflow models to pytorch HOT 3
- Does inference need its own config file? HOT 1
- data problem HOT 3
- Make model architecture compatible with deployment HOT 9
- Training error without augmentation HOT 1
- Training on gpu-c-1 scratch space HOT 7
- Normalization statistics don't need to be stored with additional key HOT 1
- Unexpected behavior in model batch prediction? HOT 8
- Runtime error testing lightning pipeline HOT 5
- unify preprocessing CLI with lightning CLI HOT 1
- Brain_2.5DUNet HOT 4
- Paper Figure 5. HOT 3
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 microdl.