search

OthersideAI / self-operating-computer

A framework to enable multimodal models to operate a computer.
https://www.hyperwriteai.com/self-operating-computer
MIT License
8.21k stars 1.09k forks source link

[FEATURE] Learning Process #174

Open MirzaAreebBaig opened 4 months ago

MirzaAreebBaig commented 4 months ago

If there is some learning process before the actual task it would be working accurately rather than navigating to unnecessary places or clicking on to wrong options.

Like AppAgent which is built for smartphone has a human intervention with learning feature which lets the user to navigate and show how the task is done then it acts upon that learning.

This would make this more faster for repetitive tasks and hence improve the flow reducing time and errors while completing tasks

Thank You :)

retsamcam commented 4 months ago

Could we just fork the project and replicate what AppAgent has done but for desktop rather than strictly android @MirzaAreebBaig ?