GDevCon North America 2025/LabVIEW Autocomplete: Difference between revisions
Created page with "{{infobox |category=presentation |name=LabVIEW Autocomplete |icon=GDevCon-NAButton.png |presentation-conference=GDevCon North America 2025 |presentation-presenter=Fadil Eledath }} {{presentation |title=LabVIEW Autocomplete |presenters=Fadil Eledath |abstract=By converting LabVIEW block diagrams into directed graphs and then Markov chains, we can make estimates of what blocks we most expect to follow a sequence of preceding blocks. This simple principle yields results th..." |
add link to github |
||
| Line 18: | Line 18: | ||
*[https://www.gdevconna.org/presentations2025 2025 Presentations - GDevCon N.A.] | *[https://www.gdevconna.org/presentations2025 2025 Presentations - GDevCon N.A.] | ||
*[https://discord.com/channels/1334339941165170799/1362523292921102547/1398052335971729450 Slides] | *[https://discord.com/channels/1334339941165170799/1362523292921102547/1398052335971729450 Slides] | ||
*[https://github.com/fadilf/DMC-Complete GitHub Repository] | |||
[[Category:GDevCon N.A. 2025]] | [[Category:GDevCon N.A. 2025]] | ||
Latest revision as of 08:35, 26 August 2025
 | |
|---|---|
| Conference | GDevCon North America 2025 |
| Presenters | Fadil Eledath |
LabVIEW Autocomplete by Fadil Eledath
By converting LabVIEW block diagrams into directed graphs and then Markov chains, we can make estimates of what blocks we most expect to follow a sequence of preceding blocks. This simple principle yields results that are surprisingly powerful and useful in situations where more complex AI models may not be ideal. In my presentation, I'll demonstrate the process I used to create graph representations of code, how I indexed them into a simple model, and the pros and cons of such an approach. I'll also provide a live demo of the model in action.