Hacker News new | past | comments | ask | show | jobs | submit | eurekin's comments login

Very impressive!

I've never looked into RC builds before - how are they controlled? Do you connect the PS4 controller directly to the esp?


Yes, the ESP32 has Bluetooth support and there is an open source library to read the inputs from a PS4 controller. So I just wrote a small program to connect to the controller and send the right commands to the motors.

I think "traditional" RC people use a separate receiver module that links to their remote instead of doing this in software.


In my case it went from a patchy 1mbps to a stable 20mbps right away, according to my router's admin page - right after resoldering the resistor and connecting the antenna.

Resoldering the 0 ohm resistor was like an adventure, but somehow managed to connect the correct pins. The end result looks like following:

https://imgur.com/a/LJflZ80


I have similar set-up - can you help out with running it? Was it in ollama?

EDIT: It seems that original authors provided a nice write-up:

https://unsloth.ai/blog/deepseekr1-dynamic#:~:text=%F0%9F%96...


Yep that's pretty much what I did, their calculation for the layers was slightly off though, I found I could offload an extra 1-2 layers to the GPUs


Oh yes I reduced it by 4 for just in case :) I found sometimes the formula doesn't work, so in the worst case -4 was used - glad at least it ran!


Can context be split on multiple GPUs?


Not my field, but from this[1] blog post which references this[2] paper, it would seem so. Note the optimal approach are a bit different between training and inference. Also note that several of the approaches rely on batching multiple requests (prompts) in order to exploit the parallelism, so won't see the same gains if fed only a single prompt at a time.

[1]: https://medium.com/@plienhar/llm-inference-series-4-kv-cachi...

[2]: https://arxiv.org/abs/2104.04473


This video cleared up my confusion and corrected my misconceptions, giving me enough knowledge to hold a one-hour discussion with an actual Toyota mechanic.

Highly recommended!


I'm not sure this is the case, but to play devils advocate - Sony themselves might have to license it from a third party


There is no reason to bend over backwards to make up excuses for bad behaviour.


It's just experience from a video monitoring project we made for a Telco operator - we had to pay up quite a few JPEG related tech, just to get a certification. Even despite that tech being free.

Historically video related field was one of the most patent and license encumbered. That's why AV1 exists.


Right there with the infamous $18 BMW heated seats subscription


The heated steering wheel software unlock on my X1 is a more reasonably priced £200.


Also context size significantly impacts ram/vram usage and in programming those chats get big quickly


Was very surprised it wasn't mentioned


That's exactly what everybody advised me against doing - finetuning on own projects. Got really discouraged and stopped. So glad someone has done it!


Almost no one knows if a project/business idea will be successful or not, so it's not much use asking. It's more productive to ask smart, experienced people how to best validate and execute an idea. People generally give useful and actionable feedback based on their experiences. Just make sure you understand who you're talking to when evaluating someone's advice.


"understand who you're talking to when evaluating someone's advice." Good you mentioned this, found out to this is a crucial part as well: Always perceive the advice you get depending on that person's background and interests (e.g. your target group, or domain-foreign expert).


> That's exactly what everybody advised me against doing - finetuning on own projects

Why would someone advise against it? IMHO that sounds as the end game to me. If it weren't so darn expensive, I'd try this for myself for sure.


I think that people suggest RAG, also because the models develop so fast that very probably the base model you finetune on will be obsolete in a year or so.

If we are approaching diminishing returns it makes more sense to finetune. As the recent advances seem to happen by throwing more compute to CoT etc maybe the time is close or has already come.


What's CoT?


Chain of Thought. When I see people using abbreviations like this I sometimes jokingly wonder what they do with all this time they're saving.


There are so many chain types it is easier to do the abbreviations. Basically extend a RAG to have a graph to influence how to either critisize itself or perform different actions. It has gotten to the point where there are libraries for define them. https://langchain-ai.github.io/langgraph/tutorials/introduct...


Perhaps they're preemptively reducing several tokens into one, for the machines' benefit.


I post in twitter and invest in crypto


Fine tuning to a specific codebase is a bit strange. It's going to learn some style/tool guidance which is good (but there are other ways of getting), at the risk of unlearning some generalization it learned from looking at 1,000,000x more code samples of varied styles.

In general I'd suggest trying this first:

- Large context: use large context models to load relevant files. It can pickup your style/tool choices fine this way without fine tuning. I'm usually manually inserting files into context, but a great RAG solution would be ideal.

- Project specific instructions (like .cursorrules): tell it specific things you want. I tell it preferred test tools/strategies/styles.

I am curious to see more detailed evals here, but the claims are too high level to really dive into.

In generally: I love fine tuning for more specific/repeatable tasks. I even have my own fine-tuning platform (https://github.com/Kiln-AI/Kiln). However coding is very broad. Good use case for foundation large models with smart use of context.


Other people have spent a lot of time on it and gotten nowhere, so I suspect there is some art to it.


They have? Is there a write up about that?


Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: