search

Garrafao / durel_system_annotators

3 stars 0 forks source link

What is the difference between xl_lexeme_annotate.py and lexeme.py? #34

Closed Garrafao closed 8 months ago

Garrafao commented 8 months ago

What is each conceptually doing? Can we merge them or remove one of the scripts? If we need to keep both, can we rename them to make their roles clear? @shafqatvirk @AinaIanemahy

Garrafao commented 8 months ago

lexeme.py seems to contain only some helper functions. It only makes sense to keep them in this separate script if they need to be accessed by other scripts, too. Otherwise they should be merged.

shafqatvirk commented 8 months ago

Yes, lexeme contains the functions to annotate the individual instances. It could be merged but also kept separate, so that things like prediction_type (and binary vs multi threshold) can be kept separate and implemented in lexeme.

AinaIanemahy commented 8 months ago

See #35 . It might make more sense to merge those two files but keep lexeme.py separate.

AinaIanemahy commented 8 months ago

I have merged random_annotate.py and xl_lexeme_annotate.py into one script annotate.py that executes the whole annotation process. lexeme.py is called from within this script and handles the aspects that only concern xl-lexeme. 5031ea3

Lexeme.py should be renamed.

shafqatvirk commented 8 months ago

this means based on the annotator, 'XL-Lexeme-Binary', 'XL-Lexeme-Multi-Threshold', 'XL-Lexeme-Cosine' should be implemented in annotate.py?

shafqatvirk commented 8 months ago

'XL-Lexeme-Binary', 'XL-Lexeme-Multi-Threshold', 'XL-Lexeme-Cosine' has now been implemented in the annotate.py so the above question is not relevant any more, so we close this issue.