Implementation: Linguistic Alignment (e.g., WHAM)

[xehu]

WHAM paper: https://aclanthology.org/P16-1050.pdf

Example application: https://web.stanford.edu/~amirgo/docs/alignmentatwork.pdf

R code: https://github.com/langcog/disc_align

---

• [ ] Copy the Template. Go to the Feature Wiki and Copy/Paste the Feature Template into a new page.

• [ ] Fill out the Template. Fill out the basic information for the feature in the template. Use the template to document your plan for implementation and major design decisions; if anything changes along the way, update the documentation as you go.

• [ ] Create a new Feature file. Create a new file in the folder src/features. The name of the file should be NAME_features.py, where NAME is the name of your feature.

• [ ] Update the entery for your feature in feature_dictionary.py. You can find the dictionary inside the src/ folder.

• [ ] Code your feature. Write the code for the feature.

• [ ] Evaluate/Unit Test. Come up with a method of evaluating your feature. How do you know that the feature is 'correct?' For simple features, this may be trivial; for more complex ones, you may need to break down each sub-function and unit test them separately, or validate on external data. Add your unit tests to the tests/ folder.

[xehu]

@pradnayapathak next step: share to @sundy1994 the code for LSM so far (you can either send via Slack or make a new branch on GitHub and put it in the src/team_comm_tools/features/ folder

[sundy1994]

Nov 7th update: Pradnaya and I met today. She'll clean up the LSM feature file and upload to a new branch, and we'll both read through the WHAM paper and meet again. Question: It seems we need some results from the current feature builder to compute the LSM score. What is the best way here?