Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Read object features table #84

Closed
wants to merge 11 commits into from
Closed

Read object features table #84

wants to merge 11 commits into from

Conversation

emilmelnikov
Copy link
Member

@emilmelnikov emilmelnikov commented May 12, 2021

Add the ability to read an object features table from object classification workflow.

Also, refactor and simplify all workflows and associated demos.

Closes #56.

* Use special "MESSAGE" visibility for providing path examples

* Remove value auto-loading from persistent storage because
  it interferes with programmatic options usage

* Simplify field labels wording

* Change class name to be more consistent with built-in ImageJ plugins
* Use new workflow classes

* Describe the demos in a base class comment

* Remove tracking demo because example files are missing from resources
Copy link
Contributor

@wolny wolny left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is very nicely done, great job @emilmelnikov! The code is much cleaner than what it was before and I'm glad that the workflows can finally be implemented using the @Plugin annotations, instead of the oldschool GenericCommand (?).
👍 :shipit:

@k-dominik
Copy link
Contributor

k-dominik commented May 27, 2021

Are these changes backwards compatible (e.g. will old macros still run)? I've tried some of my old ones, but so far without success.

for Pixel Classification it seems that parameters have been renamed:

  • projectfilename -> project
  • inputimage -> rawdata

I know this sucks, I like the more expressive parameter names, too, but we can't break old macros because of an internal refactor

@emilmelnikov
Copy link
Member Author

Are these changes backwards compatible (e.g. will old macros still run)? I've tried some of my old ones, but so far without success.

for Pixel Classification it seems that parameters have been renamed:

  • projectfilename -> project
  • inputimage -> rawdata

I know this sucks, I like the more expressive parameter names, too, but we can't break old macros because of an internal refactor

Good call, I forgot about macros.

@k-dominik
Copy link
Contributor

Just to add now that I looked at the code it is really nice to see reduced amount of duplication and boilerplate for the workflows! I tested generating the tables and this works nicely as well. We'll have to add documentation for this though, as one has to configure the csv export in ilastik first - otherwise it won't work. The real power of the table can be unleashed once we allow for exporting the Object Identities. That way (I think) one could use feature table as colormaps (somehow) :D. We should maybe ask @tischi how to best do it and provide an example.

@emilmelnikov
Copy link
Member Author

emilmelnikov commented May 31, 2021

We'll have to add documentation for this though, as one has to configure the csv export in ilastik first - otherwise it won't work.

Would this work?

image

@emilmelnikov
Copy link
Member Author

emilmelnikov commented May 31, 2021

The real power of the table can be unleashed once we allow for exporting the Object Identities.

I presume we need to change table export in ilastik for this?

@k-dominik
Copy link
Contributor

The real power of the table can be unleashed once we allow for exporting the Object Identities.

I presume we need to change table export in ilastik for this?

no, we'd need to allow exporting a different source via --export_source=Object Identities" command line argument.

We'll have to add documentation for this though, as one has to configure the csv export in ilastik first - otherwise it won't work.

Would this work?

image

that's perfect :) users don't even need to look into some docs.

@emilmelnikov
Copy link
Member Author

we'd need to allow exporting a different source via --export_source=Object Identities" command line argument.

Should we address this in a separate PR?

@emilmelnikov
Copy link
Member Author

Superseded by #106.

Object features table could be addressed in a separate PR.

@emilmelnikov emilmelnikov deleted the add-tables branch September 22, 2023 12:31
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Better output for object classification
3 participants