Fakes Guide to make Fakes. AlohaGTR

hi bro, I installed both automatic and template and copied your settings, but when I click generate an error happens:
RuntimeError: cuDNN error: CUDNN_STATUS_NOT_INITIALIZED

In the command prompt when I double click on the webui-user.bat it says:
Found GPU0 NVIDIA GeForce GT 740 which is of cuda capability 3.0.
PyTorch no longer supports this GPU because it is too old.
The minimum cuda capability supported by this library is 3.7.

Can you help me?
 
E
enforca-errado commented
how can i use the online version? Do i have to delete everything ive downloaded and do other process, or i just need to do something whit i already have?
 
D
Deleted member 4314419 commented
watch the video at the beginning of the post
 
E
enforca-errado commented
The "path to the model" step its different here, cant find the aloha69 file ;-;

Sem-tituloc431249035a383aa.md.png

sorry for bother you again bro, tryng for yours use automatic, really appreciate the help
 
D
Deleted member 4314419 commented
Did you download the 2 files and upload it to your google drive?

you have to make a folder inside your google drive and there upload the 2 files you downloaded
 
Last edited:
Feb 17, 2023
7
2
For some reason when I link my gdrive the folder doesn't appear on the left
 
3 Comments
D
Deleted member 4314419 commented
Did you download the 2 files and upload it to your google drive?
 
D
Deleted member 4314419 commented
you have to create a folder in your google drive and there you have to upload the 2 files you downloaded, or it will not appear on the left
 
peep0
peep0 commented

I think the issue was I linked my drive before adding the files and it would not update, had to switch accounts
 
I’m doing everything right I think. But it’s not generating. I assume after I input everything you’re said I press the generate button. But then nothing happens?
 
2 Comments
4
4AcODMT commented
Does the status/command window have any info?
 
E
Eyesupguardian commented
All good I sorted it out. I think It was running a little slow and i a little inpatient
 
Dec 31, 2020
177
10,365
So you trained that model? Any chance of a safetensor version?

I have been using URPMv1.3.inpainting
 
2 Comments
D
Deleted member 4314419 commented
In the next one that I will train I will create it in safetensor. In a few days I will upload it.
 
D
Deleted member 4314419 commented
that model is pretty good, try using it with sampling method DPM adaptive it will give you more realistic results
 
works great, took a while to get the folder w the 2 files to show up on the left tho

a question, can i just leave it on? or do i have to do the whole steps from the start, for example, if i reboot my pc etc?
 
Last edited:
D
Deleted member 4314419 commented
uploading your files to gogole drive only has to be done once. now you have to do the other steps but it is faster since it is installed on your google drive
 
Z
zev318 commented
do you know how long it takes to refresh? i cant use it anymore, think i used it too much haha
 
D
Deleted member 4314419 commented
I don't remember how long, but you can create another google account and do the same process haha
 
A
Acidroom08 commented
seems to be random. But when it happens to me I go to my google drive and clear out the folder of outputs and also close the page hosting SD and start over, that usually fixes it for me
 
Anyone run into the problem of not being able to mask the areas on inpaint?
Or when it does and you go to generate the image it comes up with error code “image does not match”

I had it all running so smooth the other day and now it’s just been playing up.
 
1 Comment
S
Sage4736 commented
your image is in webp format right? change it to jpg or png and try again
 
Anyone else have an issue in the webui where the generate button does nothing at some point?
 
3 Comments
Z
zev318 commented
i get that sometimes. sometimes its when i click on generate and its actually running but doesnt say it is, and usually i wait a bit and the generated picture loads.

and then a few times i just had to check that i'm still connected
 
D
Deleted member 3089319 commented
there is a github repo for automatic1111 to download it and run it on your machine, so far using it has been good. It's pretty easy to drop in the model too since you don't have to upload to gdrive
 
A
alektrafacella commented
think i need a 'step by step' to get this running with my PC/GPU (i've already exhausted my google colab allowance)
 
Aug 23, 2022
3
2
I used it yesterday and it was amazing! However today I can't get it to run. I get this error today "Warning: caught exception 'No CUDA GPUs are available', memory monitor disabled". Any idea what I am doing wrong today?
 
Comment
Can you help me? Can you send Aloha div? Cause, if I take the "model" packig, have an error...
 
2 Comments
H
Hegedus Titusz commented
No checkpoints found. When searching for checkpoints, looked at: - file /content/gdrive/MyDrive/sd/stable-diffusion-webui/model.ckpt - directory /content/gdrive/MyDrive/sd/stable-diffusion-webui/models/Stable-diffusion - directory /content/gdrive/MyDrive/aloha2 Can't run without a checkpoint. Find and place a .ckpt or .safetensors file into any of those locations. The program will exit. Fizetős Colab-termékek
 
D
Deleted member 3089319 commented
did you upload the model to your google drive and put the path to the folder where it says path to model?
 
I tried it a few days ago and it works.
Now I tried it again, but when I get to run the Start Stable-Diffusion step, I got this error :

Traceback (most recent call last):
File "/content/gdrive/MyDrive/sd/stable-diffusion-webui/webui.py", line 7, in <module>
from fastapi import FastAPI
ModuleNotFoundError: No module named 'fastapi'
 
D
Deleted member 4314419 commented
also delete it in the trash tab
 
keanurefresh
keanurefresh commented
D
Deleted member 3570765 commented
Give a day or so. It should let you again. Also try to get yourself another google account. I alternate with two accounts and had no problems so far
 
D
Deleted member 4314419 commented
try the new method, it is at the beginning of the post