[Bug]: No module named 'LDSR' #1497

smilerz · 2022-10-10T13:20:52Z

What happened?

After updating to commit 8f613bf I get the following error attempting to upscale in Image Lab (old UI, GoLatent mode).

I attempted to reclone latent-diffusion to /src/ and recopied project.yaml and model last.cpkt to /models/ldsr without success.

Version

0.0.1 (Default)

What browsers are you seeing the problem on?

Chrome, Microsoft Edge

Where are you running the webui?

Windows

Custom settings

No response

Relevant log output

Traceback (most recent call last):
  File "C:\xxxxx\Miniconda3\envs\ldm\lib\site-packages\gradio\routes.py", line 247, in run_predict
    output = await app.blocks.process_api(
  File "C:\xxxxx\Miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 641, in process_api
    predictions, duration = await self.call_function(fn_index, processed_input)
  File "C:\xxxxx\Miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 556, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "C:\xxxxx\Miniconda3\envs\ldm\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "C:\xxxxx\Miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 937, in run_sync_in_worker_thread
    return await future
  File "C:\Uxxxxx\Miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 867, in run
    result = context.run(func, *args)
  File "scripts/webui.py", line 2288, in imgproc
    ModelLoader(['LDSR'],True,False) # Load used models
  File "scripts/webui.py", line 2337, in ModelLoader
    global_vars[m] = load_LDSR()
  File "scripts/webui.py", line 385, in load_LDSR
    from LDSR import LDSR
ModuleNotFoundError: No module named 'LDSR'

Code of Conduct

I agree to follow this project's Code of Conduct

The text was updated successfully, but these errors were encountered:

SharkyMoto · 2022-10-10T21:20:19Z

I'm having this issue too, but unfortunately couldnt find a solution on my own

kftX · 2022-10-12T10:38:36Z

Exact same error here, SD-GUI is fully up-to-date

Traceback (most recent call last):
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\routes.py", line 247, in run_predict
output = await app.blocks.process_api(
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 641, in process_api
predictions, duration = await self.call_function(fn_index, processed_input)
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 556, in call_function
prediction = await anyio.to_thread.run_sync(
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
return await get_asynclib().run_sync_in_worker_thread(
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio_backends_asyncio.py", line 937, in run_sync_in_worker_thread
return await future
File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio_backends_asyncio.py", line 867, in run
result = context.run(func, *args)
File "scripts/webui.py", line 2233, in imgproc
ModelLoader(['LDSR'],True,False) # Load used models
File "scripts/webui.py", line 2337, in ModelLoader
global_vars[m] = load_LDSR()
File "scripts/webui.py", line 385, in load_LDSR
from LDSR import LDSR
ModuleNotFoundError: No module named 'LDSR'

codedealer · 2022-10-12T12:20:19Z

Two things to check:

In the Model Manager in Streamlit interface check that LDSR is actually loaded; if it's not, the download button will show up.
In the Settings the path to LDSR should point to the correct location listed in the Model Manager. If not, edit it manually or just reset the settings.

smilerz · 2022-10-12T12:25:27Z

Two things to check:

In the Model Manager in Streamlit interface check that LDSR is actually loaded; if it's not, the download button will show up.

In the Settings the path to LDSR should point to the correct location listed in the Model Manager. If not, edit it manually or just reset the settings.

The model shows loaded in Streamlit. Also, when launching the gradio webui the console shows this:

Relauncher: Launching...
Found GFPGAN
Found RealESRGAN
Found LDSR

codedealer · 2022-10-12T18:42:58Z

the location for the LDSR model was changed in streamlit. It should be located in models/ldsr to be used there.

Gradio most likely uses the old path still which is in src/latent-diffusion/experiments... etc. You can test it by moving/symlinking the file there.

smilerz · 2022-10-12T18:48:16Z

the location for the LDSR model was changed in streamlit. It should be located in models/ldsr to be used there.

Gradio most likely uses the old path still which is in src/latent-diffusion/experiments... etc. You can test it by moving/symlinking the file there.

I put it in both places to rule that out as the root cause.

codedealer · 2022-10-12T20:22:57Z

Judging from the code the models are in the right place (in the models folder) and I am unable to replicate this issue locally.

kftX · 2022-10-14T12:23:49Z

the location for the LDSR model was changed in streamlit. It should be located in models/ldsr to be used there.
Gradio most likely uses the old path still which is in src/latent-diffusion/experiments... etc. You can test it by moving/symlinking the file there.

I put it in both places to rule that out as the root cause.

Can confirm I have the files in both places as well and the issue is there too. Also, just like this

the location for the LDSR model was changed in streamlit. It should be located in models/ldsr to be used there.

Gradio most likely uses the old path still which is in src/latent-diffusion/experiments... etc. You can test it by moving/symlinking the file there.

I can confirm that just like @smilerz I have the tested with files in both places, got the same error.

And just like him, when booting up SDGUI I also have:

Relauncher: Launching...
Found GFPGAN
Found RealESRGAN
Found LDSR

However I did notice that when running the upscaler, right before the error the command line logs:

Unloaded LDSR
Loaded GFPGAN
Unloaded Stable Diffusion
Unloaded GFPGAN

And then the error comes. I am using the gradio UI.

codedealer · 2022-10-14T14:19:12Z

Somebody post concrete steps to replicate this.

smilerz · 2022-10-14T14:58:26Z

Somebody post concrete steps to replicate this.

Launch webui.cmd without parameters.
Navigate to Image Lab tab
Select any image for source
Select Upscale under Processor Selection
Select GoLatent under Upscaler Selection, accept all other defaults
Click Process

Relauncher: Launching...
Found GFPGAN
Found RealESRGAN
Found LDSR
Loading model from models/ldm/stable-diffusion-v1/model.ckpt
Global Step: 470000
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type 'vanilla' with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type 'vanilla' with 512 in_channels
Running on local URL:  http://localhost:7862/

To create a public link, set `share=True` in `launch()`.
Processing images...
Unloaded GFPGAN
Unloaded LDSR
Loaded RealESGAN
Couldn't find metadata on image
Couldn't find metadata on image
Couldn't find metadata on image
GoBig upscaling will process a total of 9 images tiled as 3x3 in a total of 9 batches.
[MemMon] Recording max memory usage...

Iteration: 1/1
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 16/16 [00:24<00:00,  1.56s/it]
[MemMon] Recording max memory usage...

[MemMon] Stopped recording.

Iteration: 1/1
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 16/16 [00:23<00:00,  1.50s/it]
[MemMon] Stopped recording.

[truncated]

scripts/webui.py:2106: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
  mask_w = make_mask_image(np.arange(grid.overlap, dtype=np.float).reshape((1, grid.overlap)).repeat(grid.tile_h, axis=0))
scripts/webui.py:2107: DeprecationWarning: `np.float` is a deprecated alias for the builtin `float`. To silence this warning, use `float` by itself. Doing this will not modify any behavior and is safe. If you specifically wanted the numpy scalar type, use `np.float64` here.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations
  mask_h = make_mask_image(np.arange(grid.overlap, dtype=np.float).reshape((grid.overlap, 1)).repeat(grid.image_w, axis=1))
[MemMon] Stopped recording.
Unloaded Stable Diffusion

Traceback (most recent call last):
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\gradio\routes.py", line 247, in run_predict
    output = await app.blocks.process_api(
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 641, in process_api
    predictions, duration = await self.call_function(fn_index, processed_input)
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 556, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 937, in run_sync_in_worker_thread
    return await future
  File "C:\Users\Chris\Miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 867, in run
    result = context.run(func, *args)
  File "scripts/webui.py", line 2288, in imgproc
    ModelLoader(['LDSR'],True,False) # Load used models
  File "scripts/webui.py", line 2337, in ModelLoader
    global_vars[m] = load_LDSR()
  File "scripts/webui.py", line 385, in load_LDSR
    from LDSR import LDSR
ModuleNotFoundError: No module named 'LDSR'

kftX · 2022-10-14T18:46:05Z

Somebody post concrete steps to replicate this.

Launch webui.cmd without parameters;
Navigate to Image Lab tab;
Select any image for source;
Select Upscale under Processor Selection;
Select Latent Diffusion SR under Upscaler Selection, accept all other defaults
Click Process

Relauncher: Launching...
Found GFPGAN
Found RealESRGAN
Found LDSR
Loading model from models/ldm/stable-diffusion-v1/model.ckpt
Global Step: 470000
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
making attention of type 'vanilla' with 512 in_channels
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
making attention of type 'vanilla' with 512 in_channels
Running on local URL:  http://localhost:7860/

To create a public link, set `share=True` in `launch()`.
Processing images...
Unloaded Stable Diffusion
Unloaded GFPGAN
Traceback (most recent call last):
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\routes.py", line 247, in run_predict
    output = await app.blocks.process_api(
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 641, in process_api
    predictions, duration = await self.call_function(fn_index, processed_input)
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\gradio\blocks.py", line 556, in call_function
    prediction = await anyio.to_thread.run_sync(
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio\to_thread.py", line 31, in run_sync
    return await get_asynclib().run_sync_in_worker_thread(
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 937, in run_sync_in_worker_thread
    return await future
  File "C:\Tools\miniconda3\envs\ldm\lib\site-packages\anyio\_backends\_asyncio.py", line 867, in run
    result = context.run(func, *args)
  File "scripts/webui.py", line 2233, in imgproc
    ModelLoader(['LDSR'],True,False) # Load used models
  File "scripts/webui.py", line 2337, in ModelLoader
    global_vars[m] = load_LDSR()
  File "scripts/webui.py", line 385, in load_LDSR
    from LDSR import LDSR
ModuleNotFoundError: No module named 'LDSR'

In my case, this was attempted on a completely fresh installation of SD-WebUI, after following the tutorial here. Before my previous install randomly broke, upscaling was working fine.

ovasdi · 2022-10-15T03:26:43Z

Hi, same issue here. Getting ModuleNotFoundError: No module named 'LDSR' when in image lab (gradio ui) after trying to upscale using Latent Diffusion SR option, default settings.

I have latest changes on master. Fresh conda environment. LDSR models in both paths. ~~Even followed this advice to checkout an earlier Merge commit. None have helped so far.~~

See #1497 (comment)

codedealer · 2022-10-15T09:24:20Z

It looks to be related to the updated requirements.txt file which fails to install LDSR correctly.

ovasdi · 2022-10-19T03:26:03Z

Hi, same issue here. Getting ModuleNotFoundError: No module named 'LDSR' when in image lab (gradio ui) after trying to upscale using Latent Diffusion SR option, default settings.

I have latest changes on master. Fresh conda environment. LDSR models in both paths. Even followed this advice to checkout an earlier Merge commit. None have helped so far.

Hi again, I wanted to give an update RE this advice which actually worked for me after I did the following:

open conda shell and cd into sd-webui directory
git checkout 72fb6ffe1fc76b668c822f7a2cc0934dc7bd08af
git checkout -b edit
edit webui.cmd commenting out L60-L74
make whatever other changes and commit
remove the conda environment (ldm or ldo or whatever): conda env remove --name ldo
- to list environments: conda env list
remove z_version_env.tmp
execute webui.cmd

After it reinstalled the packages and all I was able to use Latent Diffusion SR, gradio iu.

RE: the issue itself. I dont have time to look right now but @codedealer is on to something, I think. This change may be related.

smilerz · 2022-10-19T12:41:58Z

I can confirm that 96148e8 fixes my issue.

smilerz added the bug Something isn't working label Oct 10, 2022

ovasdi mentioned this issue Oct 19, 2022

[Bug]: Error Loading LDSR after installation of model.ckpt #488

Closed

1 task

codedealer mentioned this issue Oct 19, 2022

revert LDSR installation to be editable #1549

Merged

codedealer closed this as completed in 96148e8 Oct 19, 2022

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Bug]: No module named 'LDSR' #1497

[Bug]: No module named 'LDSR' #1497

smilerz commented Oct 10, 2022

SharkyMoto commented Oct 10, 2022

kftX commented Oct 12, 2022 •

edited

Loading

codedealer commented Oct 12, 2022 •

edited

Loading

smilerz commented Oct 12, 2022

codedealer commented Oct 12, 2022

smilerz commented Oct 12, 2022

codedealer commented Oct 12, 2022

kftX commented Oct 14, 2022 •

edited

Loading

codedealer commented Oct 14, 2022

smilerz commented Oct 14, 2022

kftX commented Oct 14, 2022 •

edited

Loading

ovasdi commented Oct 15, 2022 •

edited

Loading

codedealer commented Oct 15, 2022

ovasdi commented Oct 19, 2022

smilerz commented Oct 19, 2022

[Bug]: No module named 'LDSR' #1497

[Bug]: No module named 'LDSR' #1497

Comments

smilerz commented Oct 10, 2022

What happened?

Version

What browsers are you seeing the problem on?

Where are you running the webui?

Custom settings

Relevant log output

Code of Conduct

SharkyMoto commented Oct 10, 2022

kftX commented Oct 12, 2022 • edited Loading

codedealer commented Oct 12, 2022 • edited Loading

smilerz commented Oct 12, 2022

codedealer commented Oct 12, 2022

smilerz commented Oct 12, 2022

codedealer commented Oct 12, 2022

kftX commented Oct 14, 2022 • edited Loading

codedealer commented Oct 14, 2022

smilerz commented Oct 14, 2022

kftX commented Oct 14, 2022 • edited Loading

ovasdi commented Oct 15, 2022 • edited Loading

codedealer commented Oct 15, 2022

ovasdi commented Oct 19, 2022

smilerz commented Oct 19, 2022

kftX commented Oct 12, 2022 •

edited

Loading

codedealer commented Oct 12, 2022 •

edited

Loading

kftX commented Oct 14, 2022 •

edited

Loading

kftX commented Oct 14, 2022 •

edited

Loading

ovasdi commented Oct 15, 2022 •

edited

Loading