aipl
iot_devices
aipl | iot_devices | |
---|---|---|
4 | 5 | |
120 | 3 | |
- | - | |
9.2 | 7.1 | |
7 months ago | 27 days ago | |
Python | Python | |
MIT License | MIT License |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
aipl
-
Ask HN: Tell us about your project that's not done yet but you want feedback on
AIPL is an "Array-Inspired Pipeline Language", a tiny DSL in Python to make it easier to explore and experiment with AI pipelines.
https://github.com/saulpw/aipl
When you want to run some prompts through an LLM over a dataset, with some preprocessing and/or chaining prompts together, AIPL makes it much easier than writing a Python script.
-
The Problem with LangChain
Yes! This is why I started working on AIPL. The scripts are much more like recipes (linear, contained in a single-file, self-evident even to people who don't know the language). For instance, here's a multi-level summarizer of a webpage: https://github.com/saulpw/aipl/blob/develop/examples/summari...
The goal is to capture all that knowledge that langchain has, into consistent legos that you can combine and parameterize with the prompts, without all the complexity and boilerplate of langchain, nor having to learn all the Python libraries and their APIs. Perfect for prototypes and experiments (like a notebook, as you suggest), and then if you find something that really works, you can hand-off a single text file to an engineer and they can make it work in a production environment.
-
Langchain Is Pointless
I agree, and that's why I've been working on AIPL[0]. Our first v0.1 release should be in the next few days. https://github.com/saulpw/aipl
It's basically just a simple scripting language with array semantics and inline prompt construction, and you can drop into Python any time you like.
-
Re-implementing LangChain in 100 lines of code
I also was underwhelmed by langchain, and started implementing my own "AIPL" (Array-Inspired Pipeline Language) which turns these "chains" into straightforward, linear scripts. It's very early days but already it feels like the right direction for experimenting with this stuff. (I'm looking for collaborators if anyone is interested!)
https://github.com/saulpw/aipl
iot_devices
-
Ask HN: Tell us about your project that's not done yet but you want feedback on
Wow, what a great idea for a thread!
I'm trying to pare down my personal projects to just the really exciting ones, so I don't have much, butni think the most appropriate to the thread is https://github.com/EternityForest/iot_devices
It's mean to be a cross-framework library for creating device integrations, so you can, say, write a handler for RTL SDR weather stations, and use it in a simple script up to a mega framework.
I kind of dislike the way HASS and others handle automations where they have special purpose primitives for everything that needs lots of hand written code.
I just have config entries, they must be strings, and data points, they can be strings, numbers, bytes, or objects. You can put metadata on them. There's also a few other utilities like the ability to make subdevices, and the ability to request things from the host.
There are no special subclasses, a light bulb is just a device with a brightness point.
It currently runs my security system with object detection recording, QR decoding if desired, multiple regions, motion detection without decoding every frame, and subsecond latency streaming to the browser, a nice recordings browser that can view a recording while it's being made, etc.
-
How I wrote my own Smart Home software
My HA platform project started in 2013. Every few months or so I check back to see if HA has progressed far enough that I can ditch one of the last custom apps in my life.
It's getting there. But it's not quite there yet. Last I checked the logging still saves every change, it's not easy to set up so that it will only save average/min/max over time to save SD wear.
Creating new integrations is easy but still not quite a five minute job like it is with my extension API(https://github.com/EternityForest/iot_devices)
But yet, having custom software in one's life is generally IMHO far more of a liability than an asset.
So what I actually do is just use YoLink and Google assistant for everything I possibly can, and use custom software for video recording and unusual stuff YoLink doesn't do.
I'd love to have a one size fits all "If it need automating, use this" platform, and HA seems like it's got the potential.... but just using the YoLink proprietary platform is the lazy, trouble free, super cheap way.
-
Ask HN: Who Wants to Collaborate?
I'm working on a standard for easy drop-in home IoT drivers: https://github.com/EternityForest/iot_devices
Maybe you could go one level of meta up and instead of working on reusable components, work on reusable definitions for component interfaces.
Reuse is hard because you need a bunch of glue code. But if you had, like a standard for a toolbar, that knew how to find all the ToolbarAble objects, and the shopping cart icon just showed up, etc, things would get easier.
The shopping cart could know to look for all the payment requesting components declared in your Big Project File or whatever, and everything could stay modular ish?
GitHub is already the standard place to share generic projects.
-
Home automation dashboard generator in the terminal
Source code can be found here: https://github.com/EternityForest/iot_devices
- Minimalistic framework for creating IoT reusable Python IoT device drivers
What are some alternatives?
modelfusion - The TypeScript library for building AI applications.
vanna - 🤖 Chat with your SQL database 📊. Accurate Text-to-SQL Generation via LLMs using RAG 🔄.
hamilton - Hamilton helps data scientists and engineers define testable, modular, self-documenting dataflows, that encode lineage and metadata. Runs and scales everywhere python does.
r0b0 - r0b0 is a communication system for connecting human interface device (HID) hardware and system software; an `aconnect` for anything.
multi-gpt - A Clojure interface into the GPT API with advanced tools like conversational memory, task management, and more
SeleneCMS - CMS built as a Symfony Bundle
haystack - :mag: LLM orchestration framework to build customizable, production-ready LLM applications. Connect components (models, vector DBs, file converters) to pipelines or agents that can interact with your data. With advanced retrieval methods, it's best suited for building RAG, question answering, semantic search or conversational agent chatbots.
openai-kiss - Simple shell scripts to access OpenAI API
llm - Access large language models from the command-line
code_nitro
llm-gpt4all - Plugin for LLM adding support for the GPT4All collection of models
jekyll-sqlite - A Jekyll plugin that lets you use SQLite database instead of data files as a data source.