20:10 eine tolle interface option wäre doch, wenn er dann eines der generiertenBilder nehmen würde um daraus dann eine verbesserte version img2img zu generieren. Aber wie du in 22:40 feststellst, kommt es drauf an ob ein Model darauf trainiert wurde, was es ausgeben soll. Alle bezahlbaren "T2I AI2s" scheinen Schwächen oder Stärken zu haben. Je nachdem was sie produzieren sollen. Einige sind besser für Tiere, aber auch nicht alle Tierarten . Andere eher für Architektur usw.
hi, i just installed vs code in my new laptop, and it took 3 second just for print hello world. I don't think there is a problem with my laptop since it can run a few triple A game. do you perhaps know how to solve this problem?
Is your code written in Python? If so how you installed Python, just bare Python or with some distribution of it, for example conda. Maybe it's running virtual environment first, and then executing the code in that env, sometimes loading an env can take some seconds. And maybe here is a problem that vs code runs env first which actually takes majority of those 3 sec or so. Try runnig your script, then re-run it without closing the terminal/console window at the bottom of vs code, if the second attempt runs faster, that's the case.
This was a fun way to waste a few hours! Runs perfectly on a 8GB GPU. For GPUs with more VRAM, "txt2imghd" looks promising. Too bad I can't use that one.
17:55 Amazing drawing man😂
20:10 eine tolle interface option wäre doch, wenn er dann eines der generiertenBilder nehmen würde um daraus dann eine verbesserte version img2img zu generieren. Aber wie du in 22:40 feststellst, kommt es drauf an ob ein Model darauf trainiert wurde, was es ausgeben soll. Alle bezahlbaren "T2I AI2s" scheinen Schwächen oder Stärken zu haben. Je nachdem was sie produzieren sollen. Einige sind besser für Tiere, aber auch nicht alle Tierarten . Andere eher für Architektur usw.
Mega, genau danach habe ich gesucht!!! Vielen Dank!!!
Thanks for this nice vid 😁 AI really steals attention this year, we are hoping to get more tools to have a look at!
if we havent given any dataset then the output will come ?
Thank you for including the links
wheres the conda setup
hi, i just installed vs code in my new laptop, and it took 3 second just for print hello world. I don't think there is a problem with my laptop since it can run a few triple A game. do you perhaps know how to solve this problem?
Is your code written in Python? If so how you installed Python, just bare Python or with some distribution of it, for example conda. Maybe it's running virtual environment first, and then executing the code in that env, sometimes loading an env can take some seconds. And maybe here is a problem that vs code runs env first which actually takes majority of those 3 sec or so. Try runnig your script, then re-run it without closing the terminal/console window at the bottom of vs code, if the second attempt runs faster, that's the case.
Awesome thanks.
Is that win 11 terminal on win 10? Someone needs to tell me how to do that. Nice vid btw 👍
This response is kind of late, but wanted to respond so here goes. Basically I'm pretty sure you get it from the ms store
Wtf is "CreativeML Open RAIL M" license?!?
I would never touch anything with that type of a weird license lmao
Nice!
Thx_.
Somebody with problems?, here I find one th-cam.com/video/H6mmNxynlZw/w-d-xo.html in my computer
Hi I love your videos, HELP PLZ :(, RuntimeWarning: invalid value encountered in cast Image.fromarray(x_sample.astype(np.uint8)).save
This was a fun way to waste a few hours! Runs perfectly on a 8GB GPU.
For GPUs with more VRAM, "txt2imghd" looks promising. Too bad I can't use that one.
way too long just make 1 repo that has it all ready to do simple¬! or at least make a script that can automate all this bs!
Best part is the harder it is the lesser the compition so it's OK to be complex, smart people any find any solutions