@@ -551,22 +551,21 @@ Click to read more
551
551
<!-- BEGIN_TODO -->
552
552
- turn those scripts into addons
553
553
- ### Applies to all tools
554
- - use beartype everywhere
554
+ - use toml instead of json? it allows setting comments too
555
+ - make it installable with a setup.py on pypi
556
+ - use beartype everywhere for statyc typing and code cleanliness
557
+ - do a unique class that could be used to unify all those codes
558
+ - arguments:
559
+ - name (to differentiate each children: for example "illustrator")
560
+ - string_format (can be overloaded)
561
+ - in the init, check that indeed there is a version attribute
555
562
- add an arg to include tags or not in the LLM context for a given note, as otherwise the LLM can get confused by some acronyms
556
563
- but with a regex arg to keep only the tags that match the regex. This way we can keep only a portion of them for the LLM
557
564
- store all inference in a compressed sqlite db instead of a json. It gets too large
558
565
- add check that we indeed removed all the done tags
559
566
- actually there's no need to store the "Done" tags because all important info is stored in the field
560
567
- use xml formatting for the examples
561
568
- make use of <thinking > tags too
562
- - make it installable with a setup.py on pypi
563
- - add images to illustrate the benefits of using each
564
- - do a unique class that could be used to unify all those codes
565
- - arguments:
566
- - name (to differentiate each children: for example "illustrator")
567
- - string_format (can be overloaded)
568
- - in the init, check that indeed there is a version attribute
569
- - use toml instead of json, it allows setting comments too
570
569
- tell user how much time each answer took
571
570
-
572
571
- ### Mnemonics Creator
@@ -597,7 +596,7 @@ Click to read more
597
596
- use an eval llm like in [ wdoc] ( https://wdoc.readthedocs.io/en/latest/ ) to better filer an anki query
598
597
- actually wdoc can already be used for that! Maybe it should be converted into an addon?
599
598
-
600
- - # Tagger (In project)
599
+ - ### Tagger (In project)
601
600
- always prepend tags by ankitagger: but customizable
602
601
- always sort those tags by alphabetical order
603
602
- add modes:
0 commit comments