Skip to content

Conversation

@vladd-bit
Copy link
Member

This is the first version of these tutorials. I'm pretty sure we will need more complex ones if we aim to demo the full pipeline in one go.

Missing from these tutorials:

  • type ID (TUI) relation pair filtering
  • playing with non-relations of certain type pairs

Copy link
Collaborator

@mart-r mart-r left a comment

Choose a reason for hiding this comment

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

A few things withuot going too deep initially:

  • We don't generally want to install based on the repo dev state but rather install an actual release through PyPI
    • Refering to install in 6.1
    • Seems fine (though commented by default) in 6.2
  • Would be useful to add links to the README as well
    • Similar to the other ones
  • Would be nice to be able to add this to the workflow as well
    • .github/workflows/main.yml
    • Add as another part
    • So that we know that this actually works every time we make chagnes to this or the library

@vladd-bit
Copy link
Member Author

Ok I updated everything. I put release 1.16.0 so it will pass the workflows when the new MedCAT version gets released.

@mart-r mart-r force-pushed the rel_cat_tutorials branch from 1c577bf to 4e90c3a Compare May 20, 2025 10:38
@mart-r mart-r merged commit 5a07e4d into main May 22, 2025
7 checks passed
alhendrickson pushed a commit to CogStack/cogstack-nlp that referenced this pull request Jul 1, 2025
* RelCAT tutorials v1.

* Updated RelCAT notebooks (incl html version).

* Updated README.md

* Workflow update to incl RelCAT.

* Update MedCAT requirement to 1.16

* Allow nbmake>=1.4 for compatibility with pydantic 2.

pydantic was dropped as a requirement in nbmake==1.4.3

* Make workflow run on ubuntu-24.04

* Bump all install targets to 1.16

* Updated RelCAT tutorial to include tutorial model links.

* Allow later pandas versions

* Revert "Allow later pandas versions"

This reverts commit e794964.

* Force usage of full path for model

* Add some debug information regarding issues when loading model

* Do archive unpack in pytho nrather than in a subprocess

* Unpack model in part 6.2 to a distinct path

* Undo debug output / exception handling

* Make sure to use correct paths when downloading and unzipping

* Remove unused import

* Use relative paths again for downloaded model

* Fix usage of undefined variable

* Fix loadable model path

* Add further information to exception

* Specfiy unzip path explicitly

---------

Co-authored-by: mart-r <mart.ratas@gmail.com>
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.

3 participants