 I will talk about it in a sec, it's pretty random stuff, not scripted, nothing is scripted, except ggpt, mute all the browser types, all the browser tabs can jump into it in a sec, that was a false detection, was it? I'll cover this one quickly, just in case, I'm still testing this, it's generating music out of hand gestures, we have the code for it, that we were actually trying to improve, we'll be doing more work on this in later streams, so please do, we just might do a YouTube short, out of trying to do some video, so so there's a volume up and down, and notes going, that was a false detection of a different hand, volume up and down, and the distance between the thumb, index finger, index, determine the length of the thumb, this will be, and the other hand is mapped to a different range of notes, which is more annoying, so note, note nice, currently I'll just mute this one for now, I'm just using it for my camera as well, so there's no sound, so hello hello hello, that in a sec, I'm actually doing something else at the moment, but let me know in the comments, if you have any questions, that will divert me into using whatever it is you are asking about, so just don't forget to put your questions and comments, that's what makes us going and improving, so currently there, we're still doing the phasologic, it might be a long term thing, it's actually scrolling through this EEG, it says synthetic EEG, and we'll be using phasologic and find, detect this EMG embedded in it, nobody does this funny thing of scaling, the waveform up and down, we'll expect it just to be the same amplitude throughout, it's actually start with that quickly, so put it in our window, we have essentially four windows, we have a code, the code is split into two windows, it's visual studio code, but it has the GitHub compiler chip open, because that's what we use, and it obviously has the code, we're mainly doing a HTML JavaScript, Python, and CSS, or styling membership functions, right, so we actually have the HTML, which shows the front panel, has some description at the bottom, but we'll have to correct it after making the more recent changes, yeah, don't know what's up with this, so we have this color scheme that we're using for the project, it's this color scheme here, that's from some festival in South America, but the way the structure of this stream is, we'll do introduction, okay, let's do introduction, we are working on this website, that's the front page, there's an about section, you can read all sorts of questions, we have all sorts of social media platforms, I have my personal website, and a consulting website, that doesn't do much at the moment, the main page contains these tools, yes, they are ads, so this video brought to you by university, which is not too chubby, not too bad, but those ads targeting me hopefully getting relevant ads for you as well, because we need to cover electricity cost, and this is all running on a local online server, it's the same machine from which I'm streaming at the moment, so if you want to support this project, we are also on Patreon, when I say we, it's me, ggpt, and github, gopout, at the moment, if you want to collaborate, do let me know, so currently, originally these tools were made in LabVIEW, and some of them do not work anymore, because we went open source, 3 is in 3DM, not as in no cost, a freedom project, these 4 tools currently do not work, because they were originally, as I said LabVIEW, and now we have to redo them in Python or JavaScript, preferably JavaScript, so you pay for the electricity, so all that left from these last 4 tools is the YouTube video that showed how it meant to work, but currently we are redoing it quite a bit, it might be combining those two, having an option to compare between phasologic and neural network in terms of performance, so let's jump into it, sometimes it works, sometimes it even works, so looking at the code above, I would like some suggestions, maybe do not generate the code yourself, but generate prompts that I can use with github co-pilot, because it will have the more updated version of the script, now looking at the code above that I shared, can you check a few things, first of all, when I scroll through the data, it's scaling the signal, zooming out, and when I stop scrolling it, adjusting the amplitude, so it's pretty annoying, how can we fix that, yep, it's just warming up with something simple, it's not going to time me out is it? Certainly, here are some prompts you could use with github co-pilot to get suggestions for modifying your JavaScript code, these prompts are designed to address the issue you're having with the scaling and amplitude adjustment when scrolling through your EEG data visualization, prompt to prevent scaling during scrollers copycode, github co-pilot, I need to fix a scrolling issue in my EEG data visualization, when I scroll through the EEG data, the signal is scaling or zooming out, which should not happen, the signal should remain at a consistent scale, can you provide a JavaScript function to handle the scroll events so that the scale of the signal does not change, prompt to maintain amplitude during scrolling, prompt for dealing with dynamic canvas resizing, prompt for scroll event optimization to avoid jitter, okay, just stop reading out loud, okay, should do it, each of these prompts targets a specific aspect of the problem and asks for a solution that you can implement in your JavaScript code, when working with github co-pilot, make sure to provide as much context as necessary for the best results and adjust the prompt based on the suggestions you get in the specific behavior you wish to achieve. Right, so we have a bunch of them, I don't know why, we don't want additional functions, no. I'm encountering an issue with amplitude adjustment in my EEG chart. Currently, when I stop scrolling through the EEG data, the chart adjusts the amplitude automatically, which is not desirable. I need the amplitude to remain constant and not readjust after scrolling. Could you help me adjust my JavaScript chart update function to maintain a fixed amplitude range regardless of the scroll position? Okay, everyone sounds legit, let's try it out. Ampitude adjustment is happening because of these lines. This marks that point constant, the scales. So when this marks that point as being calculated, the multiple, you fetch the entire EEG data. So where would that be? Here, clearing another Y, but okay. Well, this is the same. Is it the same? Don't need this scale to the constant fetch data if array. Okay, we just need to, so we don't need this anymore, do we? I can just do it on a different variable, substituting display EEG. It's still down side to control of five. It's only happening during scrolling. Okay, we'll leave it for now. The other main thing is, we want to display those membership functions. Membership functions. Actually, we would like to do as much as possible within the JavaScript, so I don't even know why we need so much of a Python code. Okay, moving on. The membership functions are currently not being displayed properly. We're just displaying one value and the current amplitude, current frequency and current artifact risk are set to be not available. We need to fix that. Now question, another question is, can we do most of the processing within JavaScript and use back-end as little as possible? To address the issues with the membership functions not displaying properly in the current values showing us N, A, here are some steps you can consider on prompts for GitHub co-ballot. Correct membership function display. If only one value is being displayed, it's possible that the data isn't being parsed or parsed to chart correctly. Ensure the data structure matches what the charting library expects. Prompt for GitHub co-ballot, Arduino copycode, GitHub co-ballot, my chart for displaying membership functions is only showing one value instead of the full range. I suspect the issue might be with parsing the data received from the server. Can you suggest a way to iterate over the received data array and correctly populate the chart dataset for membership functions? Update current values. For the current amplitude, current frequency, and current artifact risk, ensure that you're updating these values each time new data is fetched or the view is updated. They should be tied to the data refresh logic. Prompt for GitHub co-ballot, SQL copycode, GitHub co-ballot, I need to update the current amplitude, current frequency. And current artifact risk, text on my web page every time new EEG data is fetched or analyzed. The values are currently showing as in A. Can you provide a JavaScript function that updates these values in real time? Processing in JavaScript. Moving processing to the front end is feasible for lightweight tasks, but keep in mind that heavy computations might be better suited for backend processing, especially if they involve complex operations or large datasets. Prompt for GitHub co-ballot, less copycode, GitHub co-ballot, I want to shift most of the data processing to the client-side JavaScript to reduce backend load. The processing involves normalizing EEG data, calculating average amplitude and frequency, and applying fuzzy logic for artifact risk. How can I implement this in JavaScript while ensuring performance is not compromised? Optimize data transmission. To reduce backend involvement, you could send raw or minimally processed data to the front end and handle the rest with JavaScript. Prompt for GitHub co-ballot, less copycode, GitHub co-ballot, I need to minimize the backend's role in my EEG data application. The backend should only send raw EEG data to the front end. Could you guide me on how to set up the front end to handle normalization, frequency extraction, and fuzzy logic analysis? Remember that while you can do more on the client-side, not all operations should be moved there. Consider the complexity of the operations, the capabilities of the client's hardware, and the potential variability in performance across different devices. For operations like FFT or fuzzy logic analysis, which can be computationally intensive, it's often better to handle these server-side unless your certain the client-side resources are sufficient and you're not dealing with sensitive data. Assume those... Where's this fuzzy control from? Fuzzy control, what? It doesn't sound like it's going to work. What about this one or that one? I think I actually have to use the second one. I don't remember yet. Did something similar before? Yeah, let's just save that file. Still haven't fixed GitHub. Hey, co-ballots. Hey, GitHub. No, the GitHub, the Git. So we're still fetching from the back-end, which is not ideal. Let's file always be the same. Just to check something. Yeah, required is not defined. Cannot use a board statement. Dead. It's not defined. Right, so we have the fuzzy control. I'm just making this up, or it actually exists. Link. Is there a fuzzy control JS? I think this is referring, so the fuzzy control here referring to my Python. I'm really confused. Get rid of it for now. Yeah, there's too much stuff happening on the back-end, I think. Okay, I want to shift most of the data processing to the client side. Okay. So when I see the client side, do I have to select the whole JavaScript? It's selecting automatically 1 to 17. It's selecting code around this segment. It keeps normalizing that. Yeah, I didn't get that bit. We're moving the membership functions to JavaScript, which is good. It just made this up. It's suggesting it again, though. Like, this is a despite code. What on Earth? I'm super confused right now. Asking to shift most of the data processing to the client side. To reduce back-end load to implement JIFS normalization. You can normalize basic JavaScript operations. Okay, so we have normalized that. Already doing something similar. Display analyzing display. Yeah, this is only many displaying stuff. So now we're doing some processing as well. It's returning. Yeah, there's a lot of debugging. Fetching that. And normalizing that. Calculating average amplitude, okay. Yeah, so these things don't actually work. This stuff needs to go into the HTML, is it? Yeah, it's important quickly, but it won't work. So using this function in the process, please note that the approach will increase the load on the client. Client's device, that's okay. The other thing is it doesn't actually work. And then to a module. Yeah, don't have those. I need the module bundler. Okay, that would not work at the moment. It is turning my JS into a module. It is in a static folder. Yeah, now this will not work. So every function that requires an import have to be commented out. Right, so we are working again. We'd like to keep things intact instead of breaking them. Yes, we need to update the current values. Yeah, we don't need a new JavaScript function. Don't need to refare to GitHub Copa, that's GitHub Copa. You can update the current amplitude, current frequency in the car. Right, I'll take a risk text on my web page every time. You should always fetch the videos I currently showing. Let's try that quickly. So you have a HTML elements with IDs. I need to check those IDs. Right, that's correct. I don't have those. Yes, we do need this function. Call this function. After current values. Can we place it here for a sec? Comment that out quickly. Okay, now the values are updating. Okay, can remove that. That's working pretty well so far. So good. We haven't have a prompt. A chart for playing memory. Memory function is only showing one value. Instead of the full range. Sounds like this might be with the parsing of the data received from the server. Looks like perfect prompt. I couldn't have written anything better myself. Open the decision might be if how we are mapping the data onto the chart. The plot membership function. You're mapping data to labels. Seems like you're treating data as an array. Yeah, this is an array. Okay, can we check plot membership functions? It's not the actual function. Here's the function const labels. Yeah, we'd like to console log the data. Do we really have it? We already have the data. But let's do it again. Membership data. What crazy. It's all this nested stuff. Why? Just doing the plot membership function. To show what I'm meant to be. Selecting a when prompting copper. Let's not stretch the data. We have an object with three properties amplitude. Alpha, three scan frequency. Each of which is an array of object. Each of this object has a name property and points property. Where points is an array of arrays. I don't like array of arrays, but okay. The map is data tables. I'm sure you would need to iterate over the array objects and then over the points. So plot data. Canvas, CTX. After destroy. I'm in that out for a second. It doesn't work anyway. Destroy is the same. Then just take the whole. This is const plot data. 8228. No, actually the problem is elsewhere now. Let's do that for the whole code. Because it seems like when we changed this function we had to change something else as well. Their message is suggest that the dataset points is undefined. Care replacing. I'll make that out quickly. Place that. Looks to be the same anyway. Can get rid of it. Console. Okay, I might have to continue this next time. See you later. Bye.