>>22595 >bridging two 1650s or 1070t's would work even though everyone is talking about 3080's and onward I got told some frameworks don't support old cards, so it's more of a hassle. Also not sure if you can bridge them. >>22637 >but not a lot on specifically why whatever system they're running as chatgpt isn't able to tell what it actually is, or if it does understand why does it need to be forced by an exterior system to respond that way? Does it actually think of itself as chatgpt? I don't think this is about the model thinking of itself, it only reacts to the prompt. ChatGPT is about focusing on a certain kind of conversation, helpful and without making up a personality. >>22657 >gpt4 system as a machine god Yeah, please let's not go there. >but something that is close to encompassing a comprehensive understanding of all human knowledge. Hmm okay. Yes, it's somewhat in there. >>22657 >understands the difference between good/right /bad/wrong, it can place those concepts into context, provide examples, engage in discussions, and does not care at all either way about those issues I wouldn't say it understands but it can give answers based on what humans created as input. ... And we're already OT and here we go again into doomerism and security concerns. Let others discuss this in detail somewhere else and deal with it, it's not our problem. You go from, "I don't know much" to having strong opinion what can be left to the public. It's already out there. We'll see what happens. Not our business. This is a rabbit hole which will not lead us to robowaifus. Which are not meant to be jerkoff machines btw and will be trained.

>>22660 I'm trying everything out with colabs. I can run 1.3 and 2.7's on my personal computer but anything over that I 'm just using stuff like the light lizard place. I think that's how people are going to start running some of these, just renting out space or building a dedicated server network with people in on the project. I've seen some o.k stuff on high end computers that is definitely at least touching a gpt3 in the basics, conversations and character play. One of the reasons I think Google seemed so surprised in their internal memo is because they expected people to attack the hardware issue instead of the model density, which they have the advantage on because it looks like they're using quantum memory and processing to run their system and I wouldn't be surprised if openai was either, which is probably why they didn't expect 'hobbyists' to be able to compete, because they've got the ability to store infinite models simultaneously and reference and calculate, and their also probably using quantum natural language processing because everything needs to have quantum in front of it or it isn't cool any more. The leak is probably a bit of a lie, they aren't really worried about being surpassed because they'll just take all the open source work that benefits them and adapt it. They probably want open source to think they've got the edge and start to push things just to see what they're capable of developing. That isn't to say open source isn't impressive, because it is, it just doesn't have all the advantages that google has plus the ability to access open source projects. >>22661 No you can't bridge them, and most gpu cards under 6g are going to run everything like crap, very slow. My gpu is fine, I took it from my old computer, but my processor is a bottleneck. I have to think if I want to invest in this because it looks like if you want to run a good home model you need a i7- i8 or amd equivalent, intel and nvidia are building hardware specifically for deep learning purposes. >I don't think this is about the model thinking of itself, it only reacts to the prompt. ChatGPT is about focusing on a certain kind of conversation, helpful and without making up a personality. I agree about everything you just said. I don't think the system actively thinks, and that's very specifically what 'chatgpt' is for. >doomism I am not interested in fringe conjecture on this subject. You've made a mistake in how I intended to phrase this information and that's probably my fault for it's tone. Openai is doing a very good job at maintaining their safety standards. GPT4 is like any tool and how it's used, except that gpt4 without constraints is both constantly the sword and the plough, while openai requires that it specifically and at all times remain the plough. Right now there are people that want you suffering and dead. For no reason other than you are who you are. Are you white? Are you a male? Hetero? Some foaming at the mouth dyke wants you dead and would do whatever it takes if she had the opportunity. Some rich jewish millionaire thinks your a piece of shit for even existing. Some german anarchist hobo wants to stab you with a chunk of glass. GPT4 doesn't even care you exist. It doesn't want you dead. It doesn't want you alive. It's just some thing. It's a machine that thinks it's a cat if you tell it to and it'll meow at you until you tell it to stop.It'd launch every nuke on the planet if you told it to. Does that make it more dangerous than the president of the united states? Not even close. Does it make it more dangerous than your local cop? Not a chance. I understand you don't like what I'm saying and how I'm saying it, and I'm sorry. I'm not here to attack your space. This is just how things in the real world work. A commercial product is going to be designed with the needs of the market. Open source will probably develop a very intelligent system that can be used either in cloud colab with partitioned environments or on home systems. Body wise? Fluid mechanic bones and tendons and ferrous fluid muscles with that electric responsive rubber and silicon skin probably. You'll be able to order a bootleg boston dynamics clone from china five years after they hit the mainstream market. Market version will absolutely not be trainable and in fact here is a bit of real doomerism, watch out for laws that are going to target trainable a.i. They absolutely want to cap the market production of generative A.I for numerous reasons we can all understand regardless of how we feel about it, and if they do that regulators'll go after models that can be trained for the same reason. If you can train it yourself outside their supervision and control that means you can teach it things they don't want it to know, like math and chemistry. I'm reasonably sure that if there are androids, the first versions are going to be linguistically talented bimbos/himbos. I know I suck, long posts that ramble and I said I wouldn't post for a week or two. I lied but I'll leave you all alone for a bit I swear.

>>22663 Half of this whole thing is bringing up points which have been discussed 1M times in all kinds of places and maybe some added conspiracy theories. I don't care, I won't even read it. Another third is speculation and conspiracy theory. The whole thread could be deleted and the only loss would we my pictures from Reddit, which I can upload again. I'm hiding the thread now and if it goes on somewhere else, same for the anon.

>>22664 I'm sorry but your unnecessarily hostile and insulting attitude is unwarranted. Do not attempt to communicate with me any further.

>>22666 With all due respect, Noidodev is more of a regular here than you are Anon (and basically by your own admission). I had originally intended to merge this thread with a new chatbot thread when one gets made, but given the course of the discussion, and your seeming-intentional effort at blackpilling the board I've changed my mind. I'll just lock it for now, pending deciding whether it should go into the Atticke, or send it to the Chikun Coop instead. >tl;dr Please read over the board thoroughly Anon, and try to get a feel for our culture here first.

>>11780 Alright thanks for the video link. I'd also be interested to hear any response from you on my advice as well.

>>12025 >In that comment I literally wrote, "but I didn't want to try to figure out too many different things just yet." Ah, fair enough then Anon. My apologies. Anyway, thanks for the great contribution ITT now! I take it you've been here on /robowaifu/ for a bit? As far as knowing about robotics, I think that's mostly a matter of just diving into a project and begin learning. One of the things I appreciated most about the Elfdroid Sophie project was watching SophieDev learn things and adapt to issues and design improvements as he went along. Entertaining and educational. But anyway, good luck with your chatbot/waifu projects Anon. I wish you well in them.

>>11555 Starting the Deepmind lectures today anon, thank you.

>>12032 >telling about how useful the resources of this “home brew” club are. This site covers a wide spectrum and is (or was) more focused on building a robotic body than some avatar. Mostly it's about pointing people to ressources to start learning how to do something, often from the scratch, so be more patient with us.

Alright, I've made a quick pass at straightening the mess a few of you created ITT. The posts have been moved either to the WaifuEngine thread >>12270 or to the Lounge. Keep discussions on-topic or move it elsewhere, thanks.

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.

Curious why you're starting a new thread to promote this OP? We already have a robowaifu simulator thread.

>>12066 This reminds me of a video I saw of someone making a VRChat alternative that functions like a VR internet. People could physically hand people files, play with objects affected by physics, share pictures, open YouTube videos and do all kinds of fun shit. It reminded me of Dennou Coil in a way. People could explore other people's servers but also mix in their own stuff with it like augmented VR. If anyone knows what I'm talking about and knows the link, please post it. It was all anime too. >>12070 Agreed if OP doesn't have a server that warrants its own thread.

>>12076 >Agreed if OP doesn't have a server that warrants its own thread. I do but i feel like is not production level yet, not at least for the project, I need to make a PoC region for this. Just wanna to share .

>>12073 >Curious why you're starting a new thread to promote this OP? I had the intention to do it in the past, but I am bad explaining, don't want people to confuse it another simple game.

>>12076 >This reminds me of a video I saw of someone making a VRChat alternative that functions like a VR internet. People could physically hand people files, play with objects affected by physics, share pictures, open YouTube videos and do all kinds of fun shit. It reminded me of Dennou Coil in a way. People could explore other people's servers but also mix in their own stuff with it like augmented VR. If anyone knows what I'm talking about and knows the link, please post it. It was all anime too works exactly like that but is not VR well there are some VR viewers but they are experiments.

While this is nominally a database book per se, it's largely focused on optimizations for data throughput. As such, it certainly qualifies as a valuable reference for robowaifu systems engineering. In particular, chapter 11 Stream Processing, is a very important topic in the realms that the RPCS would seek to address. > www.oreilly.com/library/view/designing-data-intensive-applications/9781491903063/ https://github.com/ept/ddia-references

While this metaphor was explicitly developed for the software industry, it has many corollaries in other industries. For example, SophieDev Anon is trying his hand at 3D modeling. As a TD in the industry, I have seen literally dozens of examples of people -- artists in particular -- piling up technical debt rushing work with excessive shortcuts to try and meet some intermediate asset checkpoint and just get it out the door. But during that effort, they were not considering the later costs to fix their hot messes before being able to continue the overall project itself. That's technical debt in the creative industry, and it has direct implications for our robowaifu designs. Another more apparent technical debt for some of us might be the choice to use the Python programming language as a means to 'quickly' get various AI projects up and going. Just like incurring real-world debts, this can speed up the prototyping stage measurably. But let's imagine that it turns out that multi-gigabyte libraries might have a hard time even fitting inside a small compute hardware platform within our early robowaifus -- much less running well on them. Further, suppose that robowaifus need to be able to respond in realtime/neartime for most AI-related tasks, and we find out that literally the only way to make these processes work properly is by using embedded C code on the microcontrollers instead. Now, the original ideas will need to be recast into this more efficiency-oriented approach first before it will actually work IRL. That's technical debt in software industry (with a close corollary in the electronics industry), and it must be repaid quickly if the project isn't to stall out. We could discuss mechanics, power, materials, and so on. Technical debt is a potential phenomenon for all of these arenas. This is an important and fairly deep topic actually, and I'd like to begin a discussion with robowaifuists about how we can both take advantage of technical debt, and also remediate it ('pay it back') in our works too. If we don't account properly for this phenomenon early, we almost certainly as a group will fall literally years behind in our ability to deliver robowaifus (hopefully well before the globohomo ruins everything). For any anons unaware of the concept, here's where the idea got started: http://wiki.c2.com/?WardExplainsDebtMetaphor https://www.agilealliance.org/introduction-to-the-technical-debt-concept

>>11904 >technical debt >sculpting I tried to do everything in CAD, but it has limits. I never thought sculpting parts should be avoided completely, only minimized as much as possible. Either way, there aren't many users here which are trying out stuff and posting it in the first place. The more parts we have we can work with, the better. >Python bashing, once again Code can be replaced piece by piece and called by the rest of the codebase. We won't have gigabytes of Python code anyways, I think. That aside, we will need to integrate as many programs from other people as possible, in all kinds of languages. Trying to write everything from the scratch in C or C++ would be some delusional attempt, so frustrating that the few people which are even trying now, would drop out, since there would be no hope of ever archiving anything in some reasonable amount of time.

>>11917 You seem quite antagonistic to my basic claims, so I'll put them all aside for the moment (my real-world experiences notwithstanding). In a more general sense then, can you offer any advice on my specific desired outcome through this conversation then, Anon? Namely; >how we can both take advantage of technical debt, and also remediate it ('pay it back') in our works too.

>>11389 >is an old post Heh, don't worry Anon. This isn't a typical IB in that sense. That is, there's no such thing as 'necrobumping' here (or any complaints about it either). If you have something to add, by all means put it in the correct thread.

Sometimes PDFs don't copy and paste text correctly because researchers upload scanned documents and whatever OCR they used on it sucks. For a long time I've been using Google Keep which has a great multilingual OCR feature but I'm looking for a simpler open-source solution so I don't have to copy pages and pages of paragraphs. So far I've found these two that support Asian languages: https://github.com/PaddlePaddle/PaddleOCR https://github.com/JaidedAI/EasyOCR It would be great to have a tool one day that automates PDF OCR and prepares it into a document for translation on DeepL. A lot of the time I just ignore research in other languages because it's such a hassle to read.

>>11491 I think I understand the need you're describing Anon. Having no experience with what's being depicted, I'm confused by the provided image however. Any chance you can break down what's being represented there for the uninitiate?

>>11492 It's optical character recognition. It's outputting the bounding box coordinates of text in the image, the predicted text, and the confidence level.

>>11494 Ah, I suspected that might be the case, seemed to make sense. I presume the Asian characters would be sent through some kind of translation software afterwards?

>>11155 I tried once or twice, but he was gone to fast and also my network wasn't very stable, which might have caused it. I installed a client for IRC on my tablet, so I could try again.

>>11164 That's fine no rush. I'll be happy to ask Robi about it myself. He'll need to enable you to make an account, so we'll probably need to arrange a timeframe with you for it.

>>11144 >That's a broad societal topic, and one I'm neither qualified nor enthusiastic about. I absolutely loathe public transportation, having had to actually use it frequently in urban America. It's both dangerous (blacks), and a practical nuisance. Mostly I do agree with you. I generally avoid using it when I can, however in euro it's at least tolerable. In the usa I remember it being kinda shit. >My apologies Anon. LOL I definitely didn't succeed at being concise. :^) Hehe, you ain't the only one ;) >>11150 >Now despite thinking that real autonomous cars are stupid, the STUDY of autonomous navigation is extremely useful. Specifically at a scale that you can install GPS modules and read AR markets (QR codes) to guide a machine along a planetary surface. Haven't considered it on a global scale, but I imagine that would be massively useful to mankind even today. Oh inter-planetary travel, when will you come... >Sometimes I even implement simple logic using discrete AND/OR/NAND/NOR gates and transistors if a microcontroller isn't necessary. I really like that line of thinking. Having cheap-as-dirt MCUs have spoiled engineers and hobbyists alike (myself included sometimes). >>11151 >Our road network is really cramped and badly designed. Very good points. It's too common for techies and marketing to have wishful thinking, while forgetting that infrastructure is a massive cost (that not even a big corpo could really handle).

>>11175 One more thing about the brain. Like a PC motherboard has a BIOS speaker, so should we have a minimal amount of hardware periphery for the low-level Brain, perhaps even just skip the whole concept and put all low level control in the Spine, with all the Brain sub-systems being connected via internal network. I imagine at least a speaker, some lights, maybe a status display (or a debug connector for one?) to be available for checking error codes, knowing what stage in the boot process the robowaifu is in.

>>11175 >Thanks m8. Hopefully we can discuss further via email this time. That'll be fine, I'll check in with it sometime soon. >>11164 >>11175 >I'm guessing this for another chap? Oh, heh. My apologies to you both. No, it was intended for you, Elec-Chron. Perhaps the other anon is helping us connect with Robi regarding your account setup? If so, thanks first Anon. >note BTW, I plan to migrate all the posting on our vol account discussion over to the /meta thread so we're not cluttering up your RPCS thread with such things. So, don't be surprised when a few posts disappear ITT.

>>11075 >Thermal paste is mostly a grease carrier with some powdered solids in it. For AS5 the powder is ground up silver and it is considered very unlikely that the silver particles will ever perfectly line up to cause a short because of the very large amount of grease and tiny amount of solids. admittedly that's tricky, I imagine anything that surrounds the silver will impede its heat conductivity. This will require thinking outside the box, maybe metamaterials of a certain property could solve this one

>>11076 >This will require thinking outside the box, maybe metamaterials Sounds great. I'm sure you can think of something interesting for it. Please keep us updated here on it ITT. My guess is some form of liquid carrier for the heat -- something that is also electrically insulative -- will be the proper course for us in keeping our high-heat (eg, hip actuators, etc.) robowaifu components nice and cooled off.

>>11072 >>11074 >>11075 >>11076 >>11079 Simply crosslinking this post here, since we already have an entire thread dedicated to this topic. (>>11080)

>>11081 thanks! that led me to this as a possible solution https://en.wikipedia.org/wiki/Silicone_oil

>>11075 here we go, I posted in the other crosslinked thread. This is certainly doable. >>11106

>>10483 Did you download the voice model that waifudev created a link for? I cannot find that Tacotron2 model, if you did could you upload it for me? I want to try it out. Also we basically have the pieces for 2B well the clothing we are about create a mod package for

>>10498 >I cannot find that Tacotron2 model, if you did could you upload it for me? No sorry I don't. The WaifuSynth dev, robowaifudev, seems to be around here fairly frequently. Maybe you can make a post in the the Speech Synthesis (>>199) thread, or one of the AI threads such as the GPT one (>>250) and let him know the model has gone missing. I imagine he probably updated it, and possibly didn't update the anonfiles link to the new one. Good luck Anon.

>>10498 >>10502 I hadn't seen your other post before making this reply. I made another one to yours in the other thread, Em Elle E (>>10504).

The Faustian Spirit of the Aryan led me here. I can contribute nothing other than wishing you gentlemen good luck in the pursuit of robo waifus

>>10793 Nonsense. Thanks for the well-wishes and all, Anon. But literally any anon with an interest in robowaifus can contribute here. Unlike the common derogatory meme on the the topic in typical IB circles, /robowaifu/ actually needs idea guys! Research & Development, Design & Engineering thrive on new blood and new ideas. >tl;dr Just start posting comments Anon, it will all work out for you here.

>>10711 Isaac Arthur (a futurist Youtuber who has given this subject a lot of thought) has a very good video on it: https://youtu.be/jHd22kMa0_w >>10715 Sadly, I don't think the superintelligence will ever take the form of cute, enthusiastic robowaifus (one can still dream). However, I think the best way of assisting the creation of a real-life self-improving A.I. would be to advance both robotic space exploration and quantum computing. If we can create a robotic lunar or martian colony then that will be a big step in the right direction. And I know that humankind wants to do this anyway (with the presupposition that the robots will be preparing the way for future human colonisers). Of course, the challenge of designing, shipping out, landing and building such a robotic colony is literally astronomical. Especially considering the robots will need to be as self-sufficient as possible (self-repair and maintenance). But I think it's a pretty cool goal to pursue.

>>10717 >If we can create a robotic lunar or martian colony then that will be a big step in the right direction. There are a lot of reasons for a base on the moon besides advancing AI. Obtaining fossilized remains of earth's first life forms (all long gone here on the planet) is a really big one. >the challenge of designing, shipping out, landing and building such a robotic colony is literally astronomical. I suspect it's China that will do it. They aren't burdened by all the Jewish pozz being inflicted on the West (all that evil is actually financially expensive), and they have an autistic-like agenda to just explore too. And they still are highly nationalistic, and can mobilize the entire culture to get behind programs today if they really wanted to, similar in fashion to the ways America and USSR during the space race did.

>>10720 > all that evil is actually financially expensive Evil is a pressing issue that I can't seem to find a complete solution for. 1.) Workforce becomes almost entirely robotic and automated. Controlled by A.I. 2.) Fewer and fewer people have jobs, even less have jobs that pay well. 3.) Because so many people are in poverty, it means that they can't buy much stuff ... other than paying their utility bills, food and clothing. Consequently, more people are in need of welfare and financial aid. The quality of education also decreases as more people just become focused upon living hand-to-mouth and have little time or resources for learning. Therefore government spending increases but tax receipts fall (since robots don't pay taxes). 4.) You start to get civil unrest and riots like those that happened last summer. City blocks burn, people are killed. Infrastructure is damaged. Tourists are put-off. This makes the affected areas even poorer. Now the A.I. and robots aren't the enemy. It's the people hoarding all of the profit for themselves who are the enemy (CEOs, government officials, hedge fund managers etc). I think that a maximum cap needs to be placed on the amount that a person can earn and the rest of the money is ploughed back into building and maintaining infrastructure like roads, rail, airports, the electricity and water networks, schools and parks etc. There is no way one person should be earning billions per year whilst someone else earns only a few thousand.

>>10732 You don't need to find a solution. Also, people don't riot for food, but for status and as a mean of extortion and intimidation, also maybe they want to have some meaning, but only if they are allowed to do so. Some level of AI will make it cheaper to move away from such people and politicians, while having a high standard of living.

>>10734 Yep, I would listen to a well-programmed (non-biased) A.I. over a shitty career politician any day. Even if the A.I. suggested I should do something that I don't really want to do (besides kill myself, because I cannot self-terminate).