Smorty [she/her]

I’m a person who tends to program stuff in Godot and also likes to look at clouds. Sometimes they look really spicy outside.

  • 42 Posts
  • 133 Comments
Joined 1 year ago
Cake day: July 5th, 2023
  • Smorty [she/her]@lemmy.blahaj.zoneOPtoGodot@programming.dev•Querying the Godot documentation website? Is that possible?
    13·
    9 days ago

    It is true that the editor has the build in class reference. I have already tried to retrieve the text from when that pops up, and I have managed to do that partly, however it doesn’t segment anything. The reason why I want a cool format like JSON or maybe YAML is that I can parse it and separate it into variables, which can then be nicely display in smaller UI elements.

    secret reason why I want to segment the docs (CW: overhyped topic)

    Okay, don’t tell this to anyone, because many people don’t want to hear anything about this topic, which is reasonable: it’s abnoxious and overhyped.

    whispering: The reason I want to segment the docs is because I want to embed them and use them as structured input for a locally running LLM for better context.

    shocked crowd sound

    I already made two posts about this on here: This one and this one.

  • Smorty [she/her]@lemmy.blahaj.zoneOPtoGodot@programming.dev•New GoPilot feature: Refactor
    31·
    20 days ago

    The feature I showcased is part of my Godot plugin “GoPilot”. I have not released it yet, however I will soon on the AssetLibrary as anfree and open source package for everyone to use…

    I am currently working on a CSV table translation feature and a GraphNode based Editory which provides an easy no-code solution for LLM inference using all sorts of different settings, prompting techniques and formatting options.

    I don’t have a link right now, but I will make a post on here once I actually release it, alongside a YouTube video with all its features.

  • Smorty [she/her]@lemmy.blahaj.zoneOPtoGodot@programming.dev•Copilot-like interface for Godot! (my plugin)
    2·
    1 month ago

    You are right in that it can be useful to feed in all of the contents in other related files.

    However!

    LLMs take a really long time before writing anything with a large context input. the fact that githubs copilot can generate code so quickly even though it has to keep the entire code file in context is a miracle to me.

    Including all related or opened GDScript files would be way too much for most models and it would likely take about 20 seconds for it to actually start generate some code (also called first token lag). So I will likely only implement the current file into the context window, as that might already take some time. Remember, we are running local LLMs here, so not everyone has a blazingly fast GPU or CPU (I use a GTX1060 6GB for instance).

    Example

    I just tried it and it took a good 10 seconds for it to complete some 111 line code without any other context using this pretty small model and then about 6 seconds for it to write about 5 lines of comment documentation (on my CPU). It takes about 1 second with a very short script.

    You can try this yourself using something like HuggingChat to test out a big context window model like Command R+ and fill its context windw with some really really long string (copy paste it a bunch times) and see how it takes longer to respond. For me, it’s the difference between one second and 13 seconds!

    I am thinking about embedding either the current working file, or maybe some other opened files though, to get the most important functions out of the script to keep context length short. This way we can shorten this first token delay a bit.

    This is a completely different story with hosted LLMs, as they tend to have blazingly quick first token delays, which makes the wait trivial.