-
Notifications
You must be signed in to change notification settings - Fork 68
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
Add conversational entity linking into REL #150
Conversation
We have no way of testing, because the data are not available
Hi @chriskamphuis and @hideaki-j I think this PR is pretty much done, could you have a look at it and let me know what you think? |
@hideaki-j Can you look at this, as you created CREL, I am not familiar with it |
I have skimmed through the changes of CREL parts, and it seems to be OK (Stef and I had a meeting about this and I have already shared this with him 😉) I will take the time to check the documentation and test etc in detail when I find time but I think it is good to move to the next step for now. |
Hi @hideaki-j @chriskamphuis, if you are happy with the changes, don't forget to press the 'Approve changes' button, so that the code can be merged :-) |
This PR implements converstational entity linking from @hideaki-j into REL.
First I refactored the CREL code (see informagi/conversational-entity-linking-2022#1). I left the code as-is as much as possible, and tried to depend on functions already in REL where possible. Then I copied all the code over into
./src/REL/crel
.I added some tests and documentation for the new CREL functions.
I also refactored the server to use fastapi/pydantic (in #147 so you can see the changes). I then branched this off #147, which makes it much easier to implement the conversational entity linking into the server. Note that the API itself did not change, I just added conversational entity linking to it. Fastapi/pydantic will make it much easier to update the API later on.
To check out the server working, see
./scripts/test_server.py
for an example.Todo
-> Revisit when Mention detection with Bert #151 is merged
Follow-up:
bert_md.py
once Mention detection with Bert #151 gets merged.