Run NER tagging - Githubissues

iangow / se_features

Linguistic features derived from StreetEvents

1 stars 3 forks source link

Run NER tagging #13

Closed iangow closed 4 years ago

iangow commented 4 years ago

Code is found here.

[x] Delete material related to "Alternative 1: UChicago RCC".
[x] References to schema bs_linguistics should be to se_features.
[x] References to directory proper_nouns should be to ner.
[x] Delete references to "Finance abbreviations".
[x] Perhaps get the latest NER tagger and test that we can run it on the MCCGR server.
[x] Update instructions accordingly.

iangow commented 4 years ago

@Yvonne-Han I just reassigned this to you. Are we done with this?

Yvonne-Han commented 4 years ago

@Yvonne-Han I just reassigned this to you. Are we done with this?

@iangow I’ve ticked what is done so far. Among the tasks, I’m not too sure what the Chicago RCC part is about.

Is this your new way of dealing with too many todos - by assigning someone else (i.e., me)? (Just joking 😄 I will take a look and reply to your other assignments/comments tomorrow).

iangow commented 4 years ago

Chicago RCC is referred to in the README, but I think the associated code has been removed. So the README might need to be adjusted.

Yvonne-Han commented 4 years ago

Chicago RCC is referred to in the README, but I think the associated code has been removed. So the README might need to be adjusted.

I see. So it’s just updating README (and I will need to adjust it anyways for other updates so it should be quick. I thought I need to remove some real code so I was a little bit unsure).

Yvonne-Han commented 4 years ago

@iangow I've updated the README, deleted materials related to Chicago RCC, and deleted/replaced the references as mentioned above (I also deleted one reference that is probably pointing to Anastasia's private repository - link here).

Also - I've removed the files that are not used when creating se_features.ner_class_alt_4 and se_features.ner_class_alt_7 on the MCCGR server (which includes alter_types.sql, create_ner_orgs.R, create_ner_to_exclude.R, ner_indexes.sql, numerical_intensity.ipynb, run_batch_4.sbatch and run_batch_7.sbatch).

If this is everything, you can close this issue.

iangow commented 4 years ago

All good. I extracted a couple of quick-to-address issues so we can close this one.