Also the locations of any other big file, please.
Recent community posts
Where does this download text2video_pytorch_model.pth to? The app keeps failing to download it after trying for 3 hours, but I can download it myself in 30 minutes. Where do I put the one I download myself? I can't find where it's adding a new file during the app's download. Thanks!
It's not the prettiest, but it creates a folder inside the Clip-App folder, and it's named the phrase you gave it. In the Clip-App folder, sort all files by date, and it'll be the top folder.
Confirmed to still happen in version 1.1. This time I didn't even change the image limit. I just hit render 3 times in a row without touching any settings.
Attempt 2 with same settings:
Attempt 3 with the same settings:
And here are the settings:
Oh good. I didn't know the Patreon version had fixes. I will have to join that. I've been doing a lot of playing with this to see how well I can control it, but I've run into another bug. This issue, is that I'm getting different results with the same seed/settings. I did three runs under the same seed, but different total images. I used the following settings on all:
Image mode: Resample
Prompt: Gambit vs Sentinel
no origin or target image
Hardcoded BQGan and Clip models
Generate Extra Images: 0
Resolution X:336 Y:496
Write Image every X frames: 50
Frames before stopping: various
And here are the images, which all have the same initial image, but are using the same seed to get predictable results (I was trying to go further on an idea when I saw it was just starting to get interesting). They are all image number 600:
Image limit 1251:
Image limit 901:
I'm using the same seed, so I should be getting the same results, but it's different each time. This makes it impossible to do some tests with some low image numbers, then extend on those tests when I find some good ones worth exploring. I have to always commit to a very high image count just in case I happen to get a good one.
Maybe this is fixed too? Either way, I'm joining the patreon. Thanks again.
I've been messing with VQGAN+CLIP all week. I got a Google Collab setup with tons of different approaches, and that's worked fine until Google cuts me off from GPU usage. Tried to set up a Jupyter notebook, and was successful to a degree, but ran into trouble when it attempted to run git, even though I have it installed and pathed. This standalone program was a lifesaver. I've had so much fun with it. Thank you for making it.
There are a few bugs though, which I'm sure you know about, but I thought I'd share just in case:
- "Generate extra images from the same text with random seeds" just puts all seeds in the same folder, overwriting all previous images (I should have tested this before letting it run overnight).
- On Windows, using "|" to separate subjects as instructed only results in a failure when it attempts to create the folder. It might be because Windows uses that character to indicate a new command? like with "| more" to let you hit a key between pages. I don't know.
- Running out of VRAM crashes the whole application, losing your settings and prompt text.
- Settings don't save, which means I have to work outside the tool in a text file, but it's also increased the chance of errors when re-inputting settings, like when I mistakenly put my "Write image every X frames" number into the "Generate extra images from the same text" field.
Thanks again for this. You've made it so simple and easy. So many people have been asking me how to do this AI image stuff, and I've sent them to nightcafe, or tried to explain how to use Google Collab, but your solution is the most straightforward for those who want to dive in and make a ton of stuff. It's fantastic.