>>5818 Thanks, that's both encouraging and inspiring.

>>5810 You still around Anon? How's the mlpack/LSTM/raylib project going? Any progress on it yet?

>>8933 Nah, haven't been around here much. Been focusing on making virtual waifus in Godot and using WebSockets to send data between Godot, PyTorch and the web browser. Right now my priority is to earn money with my projects and build a large GPU cluster with used Tesla K80s so I can advance my research. I still wanna make an AI toolkit with mlpack and raylib but now isn't the time. Also when raylib gets glTF support in 3.6 it will be much more ready for doing interesting AI projects. The main issue though is that most people lack the computing power to actually do anything useful with an AI toolkit. In a year or two though that'll change when people start dumping their unsupported 12 and 16 GB GPUs on the market in mass that can do amazing stuff for $100. We can snatch these cards up dirt cheap and use them in mlpack, and there won't be such an enormous barrier anymore for people to get into AI.

>>8954 Neat. Hope you make plenty of money Anon, you have some great ideas. This sounds like a nice event where good GPUs are available cheaply in a used market. Really glad to know you're still with us Anon.

Found this when people were criticizing that ML needs so much computer power and the big corps won't care. >Abstract : Strong empirical evidence that one machine-learning algorithm A outperforms another one B ideally calls for multiple trials optimizing the learning pipeline over sources of variation such as data sampling, augmentation, parameter initialization, and hyperparameters choices. This is prohibitively expensive, and corners are cut to reach conclusions. We model the whole benchmarking process, revealing that variance due to data sampling, parameter initialization and hyperparameter choice impact markedly the results. We analyze the predominant comparison methods used today in the light of this variance. We show a counter-intuitive result that adding more sources of variation to an imperfect estimator approaches better the ideal estimator at a 51× reduction in compute cost. Building on these results, we study the error rate of detecting improvements, on five different deep-learning tasks/architectures. This study leads us to propose recommendations for performance comparisons. https://hal.archives-ouvertes.fr/hal-03177159

>>9281 Correction: This conversation was about small and mybe dirty datasets instead of big data. https://project.inria.fr/dirtydata/

>>9281 >people were criticizing that ML needs so much computer power and the big corps won't care. Just looking over the abstract and admittedly not digging into the paper yet, there doesn't appear to be any contrary evidence to that complaint. I think both points are objectively, provably true. It's going to be our task here to find extremely efficient ways to run AI-like tasks, if we ever hope to have them operate (necessarily in realtime) onboard our robowaifus. Simple as. Never forget Big Tech/Gov has a demonstrably vested interest in making our task infeasible on modest, hobbyist-grade compute resources. >tl;dr It's up to us to make liars of them, Anon.

>>9285 >It's going to be our task here to find extremely efficient ways to run AI-like tasks, if we ever hope to have them operate (necessarily in realtime) onboard our robowaifus. I didn't think this was much of an issue but after I gave my chatbot an avatar the response delay became really noticeable with her sitting there blinking at me. Once I'm done with my current project I'm gonna seriously look into model compression for mobile systems and implementing these models in mlpack so we can run this stuff on embedded systems. Most of the pull requests for features that transformers require have been merged so it's ready to go now. Also it's such a pain in the ass waiting 2 minutes for PyTorch and Tensorflow to load. If this stuff is ever used in a robowaifu she's gonna have to take a nap for 20 minutes just to boot up. And the disk space usage grows exponentially each year for pointless features I will never have a use for. The mlpack code I've tried so far though compiles super tiny and starts up instantly so it gives some hope of having seamless realtime experiences even on low-end hardware.

>>9288 That is very comforting to hear Anon. I can write much more extensively in response to all your points in this post, but unless you'd like me to, then I'll simply leave it at 'GOODSPEED' :^)

>>3844 ROFL! Although, this is why I sometimes think it would be best if bots just communicate in math...or maybe ultra-rapid Morse code? That might be cool.

>>9288 Did you consider using AIML while the rest of the system starts? I think that's how it will be done eventually. There could be a list of comments to give while waking up, choosen randomly every time. Later maybe adding some new responses automatically, so she could pick up while booting and ask how you were doing what you planned to do while she was sleeping. That aside, why booting down at all? Or alternativly, why using only one computer? Because it's still development, okay. I think we're going to have systems which use some computers simultaneously. If one fails or needs to be rebooted the system would still be live and have the same knowledge. So there different ways to mitigate that, and it might only be a problem while working on a part of a system on one computer, not really an issue for a full build.

>>9377 I haven't used AIML before but that might be a good idea for dealing with loading times in a finished build. The main issue is just development really. Often I wanna test something in a model for 5 minutes and shut it down but waiting 2 minutes for it to start up wastes a lot of time. Even once PyTorch is loaded into the disk cache it still takes 15 seconds to load. One way I try get around this is by using Python's interactive console and Jupyter notebooks so PyTorch remains loaded, but sometimes the code I'm testing can't be imported easily without refactoring. It also takes some time loading large models but that could be fixed by using an SSD or possibly SD Express 8.0 cards in the future with 4 GB/s read speed.

>>9377 >I think we're going to have systems which use some computers simultaneously. If one fails or needs to be rebooted the system would still be live and have the same knowledge. You are absolutely right, and the future is here 4 decades ago Anon. 'Fly-by-wire' in aviation commonly has multiple, redundant, control computers running simultaneously. Usually in groups of 3 on modern aircraft (Although the Space Shuttle sported 4 different CnC systems). All the computers receive all the same inputs, all of them calculate these and (presumably) output all the same outputs. Or it is to be hoped so, at least. And that's the basic point; by having these redundant flight computers all running, they validate the common consensus by cross-checks and elections. If one of the three malfunctions, the other two kick it out until it 'comes to it's senses'. This leaves the actually not too unlikely scenario question "What happens if the two don't agree while the third is out of commission?" Thus the Shuttle's four machines on board. Additionally, it's not uncommon for highly-critical systems to require different contractors and different software running on at least one of the systems. That way if an unknown bug of some sort suddenly crops up, it's more likely the oddball system won't exhibit it. Safety-critical controls systems are both a complicated and fascinating field, and one ultimately of high importance to /robowaifu/. >>98 >>9390 >or possibly SD Express 8.0 cards in the future with 4 GB/s read speed. Neat, I didn't know about that yet.

>"Charticulator: Microsoft Research open-sourced a game-changing Data Visualization platform" >Creating grand charts and graphs from your data analysis is supported by many powerful tools. However, how to make these visualizations meaningful can remain a mystery. To address this challenge, Microsoft Research has quietly open-sourced a game-changing visualization platform. Haven't tried this myself yet, but I found this graph humorous & honest enough to make this post to keep track of the tool. > https://charticulator.com/index.html https://github.com/Microsoft/charticulator https://www.kdnuggets.com/2021/05/charticulator-microsoft-research-data-visualization-platform.html

>>10625 Okay, cool.

How do we get someone important to us to donate the use of one of these? I believe we could create some great robowaifu AI with it!!! :-DDD https://en.wikipedia.org/wiki/Blue_Gene

Does anyone have any resources on how the software integration would work? I.e., say you solve the vision piece so that waifubot can identify you as "husbandu," and you have the chatbot software so that you can talk to your waifu about whether NGE is a 2deep4u anime--how do you connect the two? How do you make it so that waifu realizes you, and says, "Hi, how's it going?"

>>12067 Is this one more of the many theoretical questions here? When building something, solutions for such problems will present themselves. Why theorize about it? And to what extend? Or short answer: Conditionals. Like "if".

>>12069 >Is this one more of the many theoretical questions here? No. Allow me to get more specific. I have an OpenCV based code that can identify stuff (acutally, I just got that OakD thing ( https://www.kickstarter.com/projects/opencv/opencv-ai-kit ) and ran through the tutorials), and I have a really rudimentary chatbot software. When I've been trying to think through how to integrate the two, I get confused. For example, I could pipe the output of the OakD identification as chat into the chatbot subroutine, but then it will respond to _every_ stimulus or respond to visual stimulus in ways that really don't make sense.

>>12067 In my experience the simplest way to think about it is like a database. You give the database a query and it gives a response. That query could be text, video, audio, pose data or anything really and the same for the response. You just need data to train it on for what responses to give given certain queries. There was a post recently on multimodal learning with an existing transformer language model: >>11731 >>12079 With this for example you could output data from your OpenCV code and create an encoder that projects that data into the embedding space of the transformer model.

>>12086 Exactly what my brain needed. Thanks anon.

This looks really interesting to me. llamafile https://github.com/Mozilla-Ocho/llamafile A standalone open source AI that can be run on may platforms including Raspberry Pi. It also can use other AI's other than the ones available for download. https://hacks.mozilla.org/2024/04/llamafiles-progress-four-months-in/ The AI software is getting better and better, smaller and smaller and more useful for local PC's. I imagine there's some way to train this. It could be a great advance to have these small AI's and then intensively train them on narrow dedicated task that we need. And no I don't know exactly how to do this yet. The tech is evolving rapidly to do so.

Open source consumer level simulated universe robot training AI, WOW!... maybe... Genesis https://genesis-embodied-ai.github.io/ https://github.com/Genesis-Embodied-AI/Genesis I heard about this from a guy who is using AI, successfully, to write software for his company. I comment on "Rebol" language every so often. he was very fond of it but has moved on. His forum http://www.rebolforum.com/index.cgi?f=home

o3 just came out and it is multiple times better than chatgpt-4. the argument that the underlying tech for current ai is not good enough is very weak. Also ai could mean anything if you dont establish a goal on what you want the ai to do you wont accomplish anything.

>>35044 tbh i have to rely on premade ai(such as nudenet in my case or pygamylion for the chat aspect) Not only because of lack of knowledge but because training ai is fucking expensive and time consuming and in the case of nudenet a lot of naked pictures.

>>35044 iirc o3 is hella expensive, for the kind of model architecture it uses. iirc for the entire ARC AGI test, it consumed $300,000 worth of compute.

>>35044 lol what even is this an ai good boy points graph, its like shitcoin marketing

>>35047 Here is a chart of ai research from 1950 until now. Nobody came up with something better during 74 years. I treat the current ai like a law of physics. There us no alternative in my mind. I get the impression some may want a blank slate fu that starts out with the mind of a toddler and learns as time goes on. I think with the current tech it might be possible but that imply collaboration cause not one single person would pull that off, which would imply that goal be set on stone and everyones efforts be directed towards that goal. Itd also imply continous real time training. The training data would be whats captured by the camera and sensors. The waifu would have its mind in a datacenter with gpus running at max 24/7. I think.I dont think its practical. what i have in mind is a sex bot that can position itself for sex acts and can chat. It can bend over, it can tilt its head back and forth, it can position itself on top of you. Its either or no transitioning between positions. The diference between that and the first is like comparing a hill to a mountain.

>>35049 Don't think I'm not interested in the erotic aspects of robowaifus but that's only a small part of the day(very small at my age). I want one strong enough to carry heavy stuff, dig trenches, haul firewood, steer a sailboat on a course I determine. I could think of a LOT of things I would want done if I had a bot to do the heavy, continuous, monotonous work for me. And eventually cook, clean dishes and house, the possibilities are endless.

>>35089 you have a sailboat? but do you have a 3d printer?

>>35091 >you have a sailboat? but do you have a 3d printer? Yes to both. My printer is resin. My sailboat needs mucho work, though I guess it could be sailed now if I slapped a little fiberglass on the rudder and hoisted the mast. I bought it because it was really, really cheap, came with a trailer and I want to do radical experimentation with it. If this works on my present 27 foot sailboat I will sell it and there's a Pearson 35 foot I want that also needs work but I don't want to do the radical stuff to it until I test it on the smaller one. "If" you can get a trailer to tow it you can get boats super cheap. They will need work but their fiberglass so they, mostly...depending, don't leak and make great tiny house type things. I want to travel around some sailing. Use solar, mostly, with diesel back up. BTW solar is plummeting in price. I saw panels at $0.38 a watt and these were sealed panels with the latest tech that could still produce power if one section was shaded. That's cheap.

Merry Christmas to all!

give me tay

>>35150 Press F <insert: Oh!? It's maintenance time already?.jpg>

>>31405 Very interesting, will do some testing on my Pi 5 with Gemma2 and return my findings. it also seems to be able to use the weights directly from ollama, which is super nice.

>>35044 >o3 just came out and it is multiple times better than chatgpt-4. the argument that the underlying tech for current ai is not good enough is very weak. Who claims that the tech is "too weak"? It's an online service, and not optimized to act like a human-like robot wife. We still need to build a framework to handle local LLMs. That said, improvements in such online services and self-hosted LLMs will make it easier getting help with research and coding.

>>35370 machine learning is compressed google autocomplete like most people think. this becomes more evident when machine learning us applied to videogames. a genetic algorithm can make a boxer player learn how to box on its own for example https://m.youtube.com/watch?v=SsJ_AusntiU&pp=ygUJYm94aW5nIGFp

>DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models [1][2] >abstract: >Mathematical reasoning poses a significant challenge for language models due to its complex and structured nature. In this paper, we introduce DeepSeekMath 7B, which continues pre-training DeepSeek-Coder-Base-v1.5 7B with 120B math-related tokens sourced from Common Crawl, together with natural language and code data. DeepSeekMath 7B has achieved an impressive score of 51.7% on the competition-level MATH benchmark without relying on external toolkits and voting techniques, approaching the performance level of Gemini-Ultra and GPT-4. Self-consistency over 64 samples from DeepSeekMath 7B achieves 60.9% on MATH. The mathematical reasoning capability of DeepSeekMath is attributed to two key factors: First, we harness the significant potential of publicly available web data through a meticulously engineered data selection pipeline. Second, we introduce Group Relative Policy Optimization (GRPO), a variant of Proximal Policy Optimization (PPO), that enhances mathematical reasoning abilities while concurrently optimizing the memory usage of PPO. They seem to go into some depth describing the optimization approaches they used to achieve the higher efficiencies with the available hardware. --- 1. https://arxiv.org/abs/2402.03300 2. https://github.com/deepseek-ai/DeepSeek-Math

AIs can now be used to interact with the physical world, not just conversation. Once this tech trickles down/is copied, we can have true AI integration https://www.wired.com/story/googles-gemini-robotics-ai-model-that-reaches-into-the-physical-world/

>>37469 Its not that difficult, you just look for keywords in the AI's response. For sophisticated LLMs you can even tell it what responses do what, then parse it out. I got chatgpt to give me a list of servopulses (in addition to giving it a limited range of specific expression keywords). Wouldn't be too hard to split that string into usable data.

>>37493 1. Could your method be replicated and downloaded? 2. I think the benefit that it combines vision and output. For example, I could buy a new red cup and table it hasn't seen before, set it on the new table, and say "pick up the red cup", and it will know what object is the red cup, and pick it up from the new table all without previous training or experience, just like a real human could.

>>37497 1) Its just basic keyword detection. Nothing really special, so here's some python code: a GPT4all example with a little string detection... I just yoinked it out of my much messier copy so it might have a few bugs but the concept is the same. from gpt4all import GPT4All model = GPT4All("mistral-7b-instruct-v0.1.Q4_0.gguf",device='gpu') system_template = "you will respond as a robowaifu" prompt_template = 'USER: {0}\nASSISTANT: ' prompttemp = 1 maxlen = 256 #max token length def stop_on_token_callback(token_id, token_string): # keep the ai from pretending to be the user if 'US' in token_string or 'ER' in token_string or 'SE' in token_string or 'USER' in token_string or 'User' in token_string: print("HEYO!") token_string = token_string.replace('US','') return False #token_string = '' else: return True with model.chat_session(system_template, prompt_template): n=0#number to keep track of # of user/AI exchanges userstuff = []#keep track of user input while True: user_input = input("> ") if user_input == "exit": #make a way to exit the program easily print("exiting program") break else: userstuff.append(user_input)#add user input to list log AIresponse = model.generate(userstuff[n],temp=prompttemp,max_tokens=maxlen,callback=stop_on_token_callback,repeat_penalty=3)#generate LLM response #this bit detects if there are keywords in the AI response print(AIresponse) if "smile" in AIresponse: print("replace this with code to move servos to smile") if "wave" in AIresponse print("replace this with code to move servos to wave" n+1 2) Usually object detection models are not limited to specific objects, but the general shapes. Otherwise it could only recognize the 1 specific item the trainer feeds the model. Have an object detection model to tell the shape, then a separate function to calculate the color and how close it approximates to the word you say "red". The advantage of things like chatgpt is many folks are feeding it info/functions. That's why I want to get SPUD's modular program running and publically available so folks can pack it full of all sorts of neat functions. But I'm still a little ways off from that, as I have other things to do before I get that all bundled up.

>"Real-Time Pose Detection in C++ using Machine Learning with TensorFlow Lite" https://blog.conan.io/2023/05/11/tensorflow-lite-cpp-mobile-ml-guide.html --- This will be useful for our robowaifus understanding her Master's movements (as well as others'). This system also utilizes OpenCV as well as TF Lite. Everything runs fast, with low energy-consumption, & also lightweight on hardware resource requirements (b/c straight C++ software).

IMPORTANT NEWS It appears that Backyard AI is depreciating the Desktop version. Luckily, it's offline, so I'm going to preserve the desktop version and put it on my website and Odysee. ------- Announcement section talking about the depreciating; To support building faster and more transparently, we will be deprecating the Desktop applications and tethering next week. All our apps (Desktop Web, Desktop Mac, Desktop Windows, Mobile Web, Mobile iOS, & Mobile Android) with local, tethering, & cloud inference have created a surface area that makes maintenance, updates, & new features extremely time-consuming. While we have invested significant effort in sharing code across platforms, it has still proven to be unsustainable for a team of one to three developers to maintain & build upon. The number of edge cases from supporting multiple apps and inference engines has spread our development team too thin. We want everyone who currently uses the Desktop apps to be able to continue using them without disruptions, so no further features will be added. Next week, we will be shipping out a final update to the Desktop applications that allows you to disable all updates, the hub, and all network requests. We will also be open-sourcing the full information specification on import/export, so that you have complete control of your data and can migrate to our Web and Mobile products or another service. We started with a Desktop app over two years ago, and the decision to stop working on it was a very difficult one. The reality is that we never found a revenue model that could sustainably support it, & cloud models have become significantly cheaper and private over the past two years. We are deeply appreciative of everyone who signed up for our paid cloud plans specifically to support the development of the Desktop app. I know this may be upsetting, but I hope you can all understand that we are a small development team, & to build something unique and valuable, we need to be more focused on what we work on.

>>38853 Man, I love having to update my manual literally hours after I finished it Luckily, I prepared for such an occurence

>>38853 Bummer. They seem to at least be trying to reach out to their users vaguely-responsibly. That's more than some can say. >>38855 Really glad to see you working proactively to deal with unwelcome events. That's the spirit, Anon! :^) Keep moving forward

>>38853 >Backyard AI is depreciating the Desktop version Uhhh...where would I find this to download and instructions???

>>38870 I'm archiving it right now. Currently I'm dealing with an unrelated Odysee issue, but I should have it ready in about a few hours

>>38871 https://odysee.com/@GreerTechother:3/backyard-archive:8?r=2RAnQD4k7R6nPoYVC32GJaXWCqK6sKmT

>>38873 THANKS! Downloading now.