lucasmpaim/clipspyxpyknow

A little comparison about clipspy and pyknow based on exercise of my class

Commands

Clipspy

Run:

docker-compose run python python ./run.py --type clipspy --track n --with-fix y

Run with Time Track:

docker-compose run python python ./run.py --type clipspy --track y

Pyknow

docker-compose run python python ./run.py --type pyknow --track n

Run with Time Track:

docker-compose run python python ./run.py --type pyknow --track y

Comparison

Run:

docker-compose run --rm python python ./comparison_time.py

And check the "img" folder with graphics

Notes:

The clipspy is not installed with requirements.txt because a environment problems with alpine linux the alternative is use a Ubuntu/Debian Image, but it's a largest image compared to Alpine

On Clipsypy version, the code will raise a

[BLOAD2] File /app/clipspy_version/knowledge_base/base.clp is not a binary construct file.

Ignore it, for fix it, we need to call a private function called: _load_text(<dir>)

Time Track:

Both time-track functions set the same fact's on base, not to be influenced by input time.