TheLastBen / fast-stable-diffusion

fast-stable-diffusion + DreamBooth
MIT License
7.41k stars 1.28k forks source link

Connection Errored Out on Paperspace #2153

Open AIShank opened 1 year ago

AIShank commented 1 year ago

Trying to do anything within the UI gives me this error, using the latest notebook/did a fresh install and I'm still getting the error.

firefox_fE92J82tQj

No errors are showing up in the cell either. firefox_tP22XeHHzp

stlaurentjr commented 1 year ago

Same problem. This is incredibly frustrating. Initially, I assumed the issue was related to the ControlNet parameters, but then I discovered that it persists even without using Automatic1111

TheLastBen commented 1 year ago

when you click on generate, do you get any error msg ?

CoqueTornado commented 1 year ago

I have same errors everywhere, some say Error:Something went wrong Connection errored out.

others appears like this arggg

CoqueTornado commented 1 year ago

I can generate txt to image or even deforums but not changing parameters if I select another tab, like 3D, there will appear errors

TheLastBen commented 1 year ago

does it display any error msg in the notebook cell output

AndresTarqui commented 1 year ago

Same thing happens to me, but i tried opening the UI on my cellphone and seems to work as usual, and nothing appears on the cell output.

AIShank commented 1 year ago

Something I noticed is that the startup cell isn't generating a gradio link, instead its this tensorbord thing. Screenshot_20230511-120007

Pretty sure the last time I ran the AI a few days ago it was using a gradio link and I wasn't getting these errors. Not sure if it's related or not but figured I'd mention it.

roadev commented 1 year ago

Same problem here and also noticed the change of link for gradio

TheLastBen commented 1 year ago

it's a local paperspace link, meaning you don't go through gradio or any other tunnel, straight from paperspace

AIShank commented 1 year ago

Is there anything here that seems like an issue? Haven't been getting any errors in the cell log but this is incredibly frustrating to deal with.

TheLastBen commented 1 year ago

add --share in the last cell right after $configf like this $configf --share

AIShank commented 1 year ago

I'm getting a gradio link now, but it's saying there is no interface running. firefox_VH41zuMmtn firefox_izmU5H0Tyw

TheLastBen commented 1 year ago

it's a gradio temporary issue

AIShank commented 1 year ago

Yeah I just checked their website, says things have been down for a bit regarding share links, hopefully this solution works when it's back up.

Thanks for the responses as always Ben.

TheLastBen commented 1 year ago

keep me informed about the functionality of the local paperspace server (without --share)

CoqueTornado commented 1 year ago

Hi Ben, I tried to delete all and reinstall it, tried the --share parameter but gradio does not work yet. Tried with some different models, still the error. It looks like it comes from the tensorflow link and gradio out. Tried another parameters like "--no-gradio-queue" but still the errors. I copy paste this just in case it can be appplied for our issue. The link comes from here https://github.com/AUTOMATIC1111/stable-diffusion-webui/issues/9074 " Two days ago, I managed to finally perform the embedding training. Among the things I did, which resulted in the training finally working, was updating Stable Diffusion from commits https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/22bcc7be428c94e9408f589966c2040187245d81 to commits https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/5ab7f213bec2f816f9c5644becb32eb72c8ffb89, updating Torch from version 1.12.1 to version 2.0.0 (for which the --reinstall-torch flag added after export COMMAND_ARGS= in the webui-user.sh file did not work, and instead I had to delete the venv folder within the Stable-Diffusion-webui main folder first, and then start the Stable Diffusion server as normal, during which startup the latest 2.0.0 version of Torch was automatically downloaded and installed), as well as installing all the libraries for PyTorch, which were indicated as missing, through Home Brew, which were: ninja, eigen, libuv, numpy, openblas, pybind11, python-typing-extensions, pyyaml, and libomp (although I am not confident whether installing those libraries indeed had an impact of making the embedding training finally working). I trained my embedding with 487 images, which I considered as not being relatively many in order to later generate good quality, realistically-looking images, although way more than 10 images, which is commonly suggested as being enough. Unfortunately, though, the resulting images generated are far from looking realistic. Instead, as you can see in the screenshots I have attached, they are brightly and vividly colored, looking somewhat cartoonish, although the unsharp blending of the colors, which are not unific and definitely bordered, makes them not even resemble typical cartoon drawings. Those images are somewhere in-between photos and cartoon drawings, being not either. I’m not sure whether or not this is due to the latest 2.0.0 version of Torch (which users are warned about causing issues with training, although running version 1.12.1, I couldn’t perform any training at all), the https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/5ab7f213bec2f816f9c5644becb32eb72c8ffb89 commit of Stable Diffusion itself, or something else that Stable Diffusion needs to operate? Is this a known issue currently occurring, and should I simply wait for the future updates of Stable Diffusion? " just my 2 cents. Thank you Ben for all!

TheLastBen commented 1 year ago

@CoqueTornado thanks, I removed the gradio queue, hopefully it solves the issue

CoqueTornado commented 1 year ago

yeah, it works now flawlessly! there is hope! the world is beautiful! listen to my pop music in the meanwhile hahah! cheers! and many many many thanks!!!

roadev commented 1 year ago

I'm still facing the same errors even using the --share flag

TheLastBen commented 1 year ago

@roadev how to reproduce the issue ?

AIShank commented 1 year ago

Gradio share links seem to be back up, but I'm still getting the same 'no interface is running' screen when using the --share command.

roadev commented 1 year ago

@roadev how to reproduce the issue ?

I started the machine on Paperspace (I'm using the pro tier) I just wanted to use the automatic1111 so I did:

Run install dependencies Download model (v1.5) Run the Test model block (using as source model just one of the models I was previously using) Everything appears as the first posted image of this issue and using the --share gives me a gradio link that points to a "No interface is running right now"

TheLastBen commented 1 year ago

is this the A1111 notebook or dreambooth ?

AIShank commented 1 year ago

For me it's the A111 notebook.

roadev commented 1 year ago

Dreambooth v1.5 Paperspace Notebook

TheLastBen commented 1 year ago

I'm trying to reproduce the error but couldn't, do not use the --share flag

roadev commented 1 year ago

Screenshot from 2023-05-16 13-46-11

for me it is the same without --share

I'm using the tensorboard link generated

roadev commented 1 year ago

Got the same result using the A1111 notebook.

TheLastBen commented 1 year ago

does it happen immediately after starting the webui ?

Jowe79 commented 1 year ago

Yesterday me too..same error get errored out.... It's frustrating... it's a ngrok error? Or code itself from notebook? Thanks

TheLastBen commented 1 year ago

@Jowe79 I removed ngrok from the notebook, are you using the latest one ?

Jowe79 commented 1 year ago

I don't know please can you share me the LAST notebook??thanks

Jowe79 commented 1 year ago

@Jowe79 I removed ngrok from the notebook, are you using the latest one ?

I dont use paperspace, only google colab pro and your notebook fast stable diffusion the last ben...but what is the last version upfated without ngrok?

roadev commented 1 year ago

@Jowe79 I removed ngrok from the notebook, are you using the latest one ?

I just copied the last code from your notebooks and it runs again, thank you!

TheLastBen commented 1 year ago

@Jowe79 you will always find the latest notebooks in the folder named "latest_notebooks", they are updated each time the first cell is ran

Jowe79 commented 1 year ago

The latest notebooks but where?? Can you give me the link?

TheLastBen commented 1 year ago

in the file explorer with the all the folders and files

Jowe79 commented 1 year ago

Sorry but i using colab notebook...in the section on the left where there are the drive folder etcetc?? Any screenshot for visual? Thanks

--Jordi--


De: Ben @.> Enviado: miércoles, 24 de mayo de 2023, 22:50 Para: TheLastBen/fast-stable-diffusion @.> CC: Jordi @.>; Mention @.> Asunto: Re: [TheLastBen/fast-stable-diffusion] Connection Errored Out on Paperspace (Issue #2153)

in the file explorer with the all the folders and files

— Reply to this email directly, view it on GitHubhttps://github.com/TheLastBen/fast-stable-diffusion/issues/2153#issuecomment-1561908634, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A7QVQDB6UQ6VTTIRTYFQU4DXHZYCNANCNFSM6AAAAAAX5WKPKM. You are receiving this because you were mentioned.Message ID: @.***>

TheLastBen commented 1 year ago

that doesn't apply to colab, for colab you just use the official link : https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb

Jowe79 commented 1 year ago

But this colab have maintenance? You'll update for fix bugs etc or get errored out desconnections?

--Jordi--


From: Ben @.> Sent: Thursday, May 25, 2023 7:32:34 AM To: TheLastBen/fast-stable-diffusion @.> Cc: Jordi @.>; Mention @.> Subject: Re: [TheLastBen/fast-stable-diffusion] Connection Errored Out on Paperspace (Issue #2153)

that doesn't apply to colab, for colab you just use the official link : https://colab.research.google.com/github/TheLastBen/fast-stable-diffusion/blob/main/fast_stable_diffusion_AUTOMATIC1111.ipynb

— Reply to this email directly, view it on GitHubhttps://github.com/TheLastBen/fast-stable-diffusion/issues/2153#issuecomment-1562296653, or unsubscribehttps://github.com/notifications/unsubscribe-auth/A7QVQDHFFOILY5ZDZRBZJSTXH3VHFANCNFSM6AAAAAAX5WKPKM. You are receiving this because you were mentioned.Message ID: @.***>