search

patham9 / NarsGPT

A NARS implemented as a GPT model prompted to invoke reasoning steps, with NARS-based memory and control machinery implemented in Python.
MIT License
40 stars 15 forks source link

Replace Term_AsSentence() with a shared parser library from the ONA project. #14

Closed kripper closed 6 months ago

kripper commented 6 months ago

(This issue is for https://github.com/opennars/NARS-GPT. Please also see https://github.com/patham9/NarsGPT/issues/13)

When input is <door --> [front]>. :|:, ground() generates embeddings for door hasproperty front>. :| And for the input <((<($1 * #1) --> isin> &/ <(nars * #1) --> isin>) &/ <({SELF} * $1) --> ^pick>) =/> <(nars * $1) --> pick>>, it generates embeddings for When '$1 isin #1 and when then nars isin #1' then '{SELF} ^pick $1' causes 'nars * $1 isa pick'

I believe the Term_AsSentence() should better use a shared library from the ONA project.