I feel it's mostly the novelty of it, how alien it looks compared to what we are used to. But for me once I've seen a few of them, it quickly wears out and becomes somewhat repetitive, oddly enough. Still, very interesting.
WOW! this blew me away, i was saying "what??" . i cant believe how seldon creative peoples are. i mean, applause, that looked like something i never seen. seen the ai pictures on youtube after the latest progress . the training for "beauty" or is it the conceptual learning, i dont know what lead to therise in AI creativity lately, but this one is most insane ive seen. sadly ive no time, but those tools i seen, man, one could really nerd out with them
As hobbyist game developer with no artistic skills I was thinking one day I would have to worry about buying a bunch of assets to try and publish something, but this opens a lot of possibilities for me... Is the code open source? Either way it is a field that I never thought of delving into which might be a pretty good tool to have...
I have to say I love the style of art and texture you were able to pull with the AI in your closing image far more than the images during the video. What are some of the keywords and artists you use?
Hi Cody. I'm just a random guy but I have an idea for you. Google has a website called "Google Arts and Culture" that you can use to explore artists and styles. I'm sure if you include the names of artists and art movements that interest you, you can generate some cool stuff. Let me know if you generate anything cool 😎 😎
Had no idea what AI art was, these ppl on IG have been blowing my mind this whole time 😂😂 Still amazing what they do, just thought it was all hand drawn
Thank you for this tutorial! Unfortunately, I got an error ""RuntimeError: CUDA out of memory...try setting max_split_size_mb to avoid fragmentation" but can't find where to edit that setting.
Yes, I should have said that you really do have to watch the size of your images, especially on a free tier account. Colab does offer a pro subscription for I believe $9.99 a month and I've found it worth it because you'll get more GPU memory. If this happens you'll want to click 'restart runtime' up in the menu at the very top under Runtime. Lower the size of your image and try it again.
Getting error on the installing dependencies step: "ERROR: Could not find a version that satisfies the requirement imagio (from versions: none) ERROR: No matching distribution found for imagio"
This is really fantastic! I believe if I can learn how to do this I will (eventually) come up with pictures/videos that are close to what happens in my mind. It would be incredible to be able to share with people what I “see/live” every night (and sometimes even in the daytime). Neuraldyvergence has been around a long time (I believe…) it’s only now that normal people can create it. Having an augmented brain from birth is both a curse and a blessing but mostly a curse. (Bipolar) It would be interesting to get a Neralink network hookup directly into my brain and feed this into Neuraldivergent AI. 1000 Thank you’s for this video!!!
I got an error: ERROR: Could not find a version that satisfies the requirement imagio (from versions: none) ERROR: No matching distribution found for imagio
Hey man i love this content it is super helpful and it inspires me to create. When do you think you would touch on creating something like the video you placed at the end? I joined sportsracer patreon but this a fairly new concept to me so i wouldnt know how to go about using the notebook! Keep up the good work.
You can change the resolution of the image in the parameter box but you don't want to go too high. If you set a huge image size the GPU will run out of VRAM and it won't be able to run the transformations. With the best cards in Collab you could aim up to 800x800 pix. You can always upscale the image generated online with another AI specifically for upscaling. It will keep that AI look while making the image bigger. Hope it helps.
Hello!! Thanks for this amazing tutorial. I am an absolute beginner who is interested in this style for a long long time. I would like to ask one thing, since I am exploring the code, I was wondering how do I put in multiple target images that I already uploaded. Do I put a "," or ";" between image directories? Do I put "/content/" before the second directory? I really like to try with multiple targets. Thank you once again!
Noob question. I've seen a couple of these AI art generators online. Are they all essentially the same thing? Or do they have different parameters that actually make them different "artists" in essence?
the training data sets greatly influence their end result, so if a GAN etc is trained on one data set and another uses a different data set to train, both will look different for sure
For some reason I keep on getting this error: NameError Traceback (most recent call last) in () 4 get_ipython().system('mkdir steps') 5 ----> 6 metadata = PngInfo() 7 for k, v in gen_config.items(): 8 try: NameError: name 'PngInfo' is not defined Can you help me with this?
i keep getting this error when running the download models - ERROR: Could not find a version that satisfies the requirement imagio (from versions: none) ERROR: No matching distribution found for imagio mkdir: cannot create directory ‘steps’: File exists
"Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them."
you do have to install packages every time you start a new session because it's connecting to a new virtual machine each time. But once your session is going you can generate many images without reinstalling. It is possible to cache some of the heavier dependencies, such as the models, in google drive.
I have been using this tool for a few months, but have never gotten a video like the one you have at the end (7:18). Can you elaborate on those settings you did for that one?
I will definitely elaborate on this. The tool that I used to create it is another colab notebook called Pytti by @sportsracer48 on Twitter. He has a patreon required for access, but I've found it worth $5 a month just based on fun alone www.patreon.com/sportsracer48/posts
@@neuraldivergentaigenerated1711 actually I just signed up! Very fun, but I was referring to settings you like (the ones you used in the video I referred to). For zooming, resolution, etc. And not text prompts per say. Thank you!
So I loaded into this, and immediately it gave me an error saying that there was a "Name Error" and it pointed to "'PngInfo' is not defined." Keep in mind, this is the first time I ever booted this up, so I touched nothing. Anyone know the solution?
I don't have nividia so it's telling me something about the GPU not connected and it creates using CPU and takes forever lol, none of the photos are ever fully finished even after 6+ hours
Finally a very good tutorial that really explains everything. Thank you so much.
Looking forward to a more advanced guide soon!!
that's what i love about ai, than you so much for this simple and quick tutorial. will try out immediately
This absolutely just shatters my mind. There's something about actual art from an AI that's inherently more interesting than an average hand drawn.
I feel it's mostly the novelty of it, how alien it looks compared to what we are used to. But for me once I've seen a few of them, it quickly wears out and becomes somewhat repetitive, oddly enough. Still, very interesting.
WOW! this blew me away, i was saying "what??" . i cant believe how seldon creative peoples are. i mean, applause, that looked like something i never seen.
seen the ai pictures on youtube after the latest progress . the training for "beauty" or is it the conceptual learning, i dont know what lead to therise in AI creativity lately, but this one is most insane ive seen. sadly ive no time, but those tools i seen, man, one could really nerd out with them
Can I run the code on my machine instead of using Google drive, and if so would a RTX 2060 handle it better than the free version of drive?
No, you need something wayyyyy more powerful than anything even available to the average consumer.
I wanted to reply to help the algorithm. I’ve been using this notebook a lot. It’s neat to learn on. Thanks!
Great tutorial ! Please continue!
Very helpful tutorial, I wish I had this when starting out!
Looking forward to the next video. 🏆
As hobbyist game developer with no artistic skills I was thinking one day I would have to worry about buying a bunch of assets to try and publish something, but this opens a lot of possibilities for me... Is the code open source? Either way it is a field that I never thought of delving into which might be a pretty good tool to have...
I have to say I love the style of art and texture you were able to pull with the AI in your closing image far more than the images during the video. What are some of the keywords and artists you use?
Hi Cody. I'm just a random guy but I have an idea for you. Google has a website called "Google Arts and Culture" that you can use to explore artists and styles. I'm sure if you include the names of artists and art movements that interest you, you can generate some cool stuff.
Let me know if you generate anything cool 😎 😎
@@semeyeify
So helpful, thanks!
Had no idea what AI art was, these ppl on IG have been blowing my mind this whole time 😂😂
Still amazing what they do, just thought it was all hand drawn
My brain just exploded in the most pleasant way. Thank you :)
Hey man...great info. Hope to see more walk-through style videos by you.
Just subd!
really cool
I had no idea this existed, my mind is blown!
Clear and fun tutorial, hope to see more.
The last part is so trippy!
WOW!!!! This is amazing! I am SO hooked!
Thank you for this tutorial!
Unfortunately, I got an error ""RuntimeError: CUDA out of memory...try setting max_split_size_mb to avoid fragmentation" but can't find where to edit that setting.
Yes, I should have said that you really do have to watch the size of your images, especially on a free tier account. Colab does offer a pro subscription for I believe $9.99 a month and I've found it worth it because you'll get more GPU memory.
If this happens you'll want to click 'restart runtime' up in the menu at the very top under Runtime. Lower the size of your image and try it again.
@@neuraldivergentaigenerated1711 Worked! Thank you very much! Can't wait for the next video! :)
I had the same issue - thanks!
Getting error on the installing dependencies step:
"ERROR: Could not find a version that satisfies the requirement imagio (from versions: none)
ERROR: No matching distribution found for imagio"
Helps a lot with depression, to me. Thanks
Thank you so much man, I’ve wanted to do this from the moment I saw first saw its implementation
This is really fantastic! I believe if I can learn how to do this I will (eventually) come up with pictures/videos that are close to what happens in my mind. It would be incredible to be able to share with people what I “see/live” every night (and sometimes even in the daytime). Neuraldyvergence has been around a long time (I believe…) it’s only now that normal people can create it. Having an augmented brain from birth is both a curse and a blessing but mostly a curse. (Bipolar)
It would be interesting to get a Neralink network hookup directly into my brain and feed this into Neuraldivergent AI.
1000 Thank you’s for this video!!!
Looks very interesting. I will have to try that. Great video
Thank you so much for this information you’re awesome. I hope you continue doing informative tutorials.
Better example for ai art. Best explane
Wow thanks for getting me started so easily!
I got an error:
ERROR: Could not find a version that satisfies the requirement imagio (from versions: none)
ERROR: No matching distribution found for imagio
Amazing video, thanks for sharing!
Hey man i love this content it is super helpful and it inspires me to create. When do you think you would touch on creating something like the video you placed at the end? I joined sportsracer patreon but this a fairly new concept to me so i wouldnt know how to go about using the notebook! Keep up the good work.
Yes will definitely cover this!
Thanks for sharing!
if any body gets errors imagio could not be statisfied or something check and edit imagio to imageio is not miss spelled.
I LOVE that first image. Could you tell me some keywords that might give me similar vibes in my images?
Thank you so much for this!
Any news on the next chapter regarding animated versions of all this
Thank you for this great intro !
Super handy, thank you!
This is amazing. Thank you.
insane, mindblowing
Thank you!
Hi, I have the following error "NameError: name 'PngInfo' is not defined" I'm using my own image and using the path name
Is there a way to render large images? (for instance using local GPU)
This is beautiful!
You can change the resolution of the image in the parameter box but you don't want to go too high. If you set a huge image size the GPU will run out of VRAM and it won't be able to run the transformations. With the best cards in Collab you could aim up to 800x800 pix. You can always upscale the image generated online with another AI specifically for upscaling. It will keep that AI look while making the image bigger. Hope it helps.
Hello!! Thanks for this amazing tutorial. I am an absolute beginner who is interested in this style for a long long time.
I would like to ask one thing, since I am exploring the code, I was wondering how do I put in multiple target images that I already uploaded.
Do I put a "," or ";" between image directories? Do I put "/content/" before the second directory? I really like to try with multiple targets.
Thank you once again!
hi! im wondering the exact same thing, and was curious if you managed to work this out? im exploring this for my art exams this year!
@@audi7700 lmao are you gonna cheat on your art exams using this ai art?
@@peling8307 naw it’ll help me in exploring the idea of what exactly art is defined as
@@audi7700 wow thats a good answer for teachers too lmao
The beat at the beginning is so hard for no reason
PLEASE PLEASE PLASE make those new videos and teach us all more!!!
cool stuff man.thank you
Noob question. I've seen a couple of these AI art generators online. Are they all essentially the same thing? Or do they have different parameters that actually make them different "artists" in essence?
same same haha
the training data sets greatly influence their end result, so if a GAN etc is trained on one data set and another uses a different data set to train, both will look different for sure
@@sprocket2cog Interesting. I thought they just all did a google search for images lol
Love the tut. However..
would you please get a better mic?
just for us that are hit with OCD 🎤😂
This is really great.
Thank you so much for this video man ❤
Wowe thanks for this!!
Tks for this tutorial. Are you planning another more advanced soon?
Good stuff!
Wow this is amazing
wow that was very good thanks
Nice!
the ending seems to imply there are subsequent, more advanced tutorials but i'm not finding them here in this channel?
7:37 amazing
Thanks! This is a very helpful video. :) These images are truly original? And, they're licensed for commercial use?
Astounding! :O
please make that tutorial on animation and ai
are you going to do more tutorials?
For some reason I keep on getting this error: NameError Traceback (most recent call last)
in ()
4 get_ipython().system('mkdir steps')
5
----> 6 metadata = PngInfo()
7 for k, v in gen_config.items():
8 try:
NameError: name 'PngInfo' is not defined
Can you help me with this?
same
same!! so frustrating!!!
Is it possible to draw something yourself and upload it as sort of a guide to generate?
Will you do another video on this?
But more advanced ofcourse
Woah, the horrible creature you went for looked like A melted and demonic Sonic 😳
beautiful
i keep getting this error when running the download models
-
ERROR: Could not find a version that satisfies the requirement imagio (from versions: none)
ERROR: No matching distribution found for imagio
mkdir: cannot create directory ‘steps’: File exists
me too dude
Great video! How would you implement this in a website?
great stuff! subd
So all of a sudden my images are coming up not found. I tried png, jpg and nothing is working. I came back to your video, I don't see anything wrong!
"Once men turned their thinking over to machines in the hope that this would set them free. But that only permitted other men with machines to enslave them."
Who owns the copyrights for the AI generated images?
People trying to make shitty NFTs:
Write that down! Write that down!
do you have to download python packages/libraries for each session or can you save those for the next time you run it?
you do have to install packages every time you start a new session because it's connecting to a new virtual machine each time. But once your session is going you can generate many images without reinstalling. It is possible to cache some of the heavier dependencies, such as the models, in google drive.
So this is the origin of HALOS!dustbelief phase3
MORE CONTENT PLEASE
this is great
What is the the thumbnail image?
Did you make it?
I like it a lot kind, of looks like its studio ghibli
Thanks this video was very good. What text prompt created the video at 7:29?
can you do a tutorial for the inifite zoom at the end?
ERROR: Could not find a version that satisfies the requirement imagio (from versions: none)
My image is generating realy slow. I have a good computer and all, what can i change?
Great Channel! But when I run a new text why is an error coming in AI image generation!
Part 2?
We want more videos
How did you do the ending video loop?
Really cool shit
I have been using this tool for a few months, but have never gotten a video like the one you have at the end (7:18). Can you elaborate on those settings you did for that one?
I will definitely elaborate on this. The tool that I used to create it is another colab notebook called Pytti by @sportsracer48 on Twitter. He has a patreon required for access, but I've found it worth $5 a month just based on fun alone
www.patreon.com/sportsracer48/posts
@@neuraldivergentaigenerated1711 actually I just signed up! Very fun, but I was referring to settings you like (the ones you used in the video I referred to). For zooming, resolution, etc. And not text prompts per say. Thank you!
@@gentrydavidson oh got it. yeah the camera settings are
"translate_x": "0", "translate_y": "0", "translate_z_3d": "300* sin(t/5*pi)**2", "rotate_3d": "[1, 0, 0, 0]", "rotate_2d": "0", "zoom_x_2d": "0", "zoom_y_2d": "0", "lock_camera": true, "field_of_view": 60, "near_plane": 200, "far_plane": 10000,
@@neuraldivergentaigenerated1711 the fact you are replying and help your community is why I shall subscribe 🥳 good video thank you
i have different version of this same code
How do we find more texts like cyberpunk forest by Salvador Dali
Do I always have to do save on drive?
thank you very much! could you please do a tutorial for the sportsracer48's collab pytti 5 beta?
Yes I'd like to get to that eventually
@@neuraldivergentaigenerated1711 thank you very much!
So I loaded into this, and immediately it gave me an error saying that there was a "Name Error" and it pointed to "'PngInfo' is not defined." Keep in mind, this is the first time I ever booted this up, so I touched nothing. Anyone know the solution?
is there a way to add negative prompts???
I followed your instructions and keep getting stuck on this error "NameError: name 'PngInfo' is not defined". Any advice?
Hi,Thanks,but I can't open the link of The tutorial notebook
I don't have nividia so it's telling me something about the GPU not connected and it creates using CPU and takes forever lol, none of the photos are ever fully finished even after 6+ hours
Is there any tutorial on how to 'dock' this in a cloud service so it can render things in batches? Where should I be searching for this?