Fooocus
Fooocus copied to clipboard
lllyasviel
[Bug]: custom preset fails
Checklist
- [X] The issue has not been resolved by following the troubleshooting guide
- [ ] The issue exists on a clean installation of Fooocus
- [X] The issue exists in the current version of Fooocus
- [X] The issue has not been reported before recently
- [ ] The issue has been reported before but has not been fixed yet
What happened?
preset fails, default or pony works fine. Suspect there is an issue with my preset started in v2.5.1
Steps to reproduce the problem
Here is my custom turbo preset Fooocus starts up fine and sees the checkpoint in the drop down. Verify it exists in the current location where it is being referenced.
($HOME is not what is actually out put, just masking my path/directories for github)
What should have happened?
comparing the default.json preset to mine, i really only see the previous models used as a glaring difference.
It should see the model/file and continue on it's merry way.
What browsers do you use to access Fooocus?
Mozilla Firefox
Where are you running Fooocus?
Locally
What operating system are you using?
mac osx Sonoma 14.5
Console logs
start up args
`[System ARGV] ['$HOME/Fooocus/entry_with_update.py', '--theme=dark', '--always-cpu', '--disable-offload-from-vram', '--listen', '--unet-in-fp8-e5m2', '--preset', 'turbo']`
--------
Using split attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using split attention in VAE
extra {'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_l.logit_scale'}
left over keys: dict_keys(['conditioner.embedders.0.logit_scale', 'conditioner.embedders.0.text_projection'])
Base model loaded: $HOME/Fooocus/models/checkpoints/DreamShaperXL_Turbo_v2_1.safetensors
VAE loaded: None
Request to load LoRAs [] for model [$HOME/Fooocus/models/checkpoints/DreamShaperXL_Turbo_v2_1.safetensors].
Fooocus V2 Expansion: Vocab with 642 words.
Fooocus Expansion engine loaded for cpu, use_fp16 = False.
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Started worker with PID 94731
App started successful. Use the app with http://localhost:7865/ or 0.0.0.0:7865
[Parameters] Adaptive CFG = 7
[Parameters] CLIP Skip = 2
[Parameters] Sharpness = 3
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] Seed = 1093821411714290983
[Parameters] CFG = 2
[Fooocus] Downloading control models ...
[Fooocus] Loading control models ...
[Parameters] Sampler = dpmpp_sde - karras
[Parameters] Steps = 8 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
Traceback (most recent call last):
File "$HOME/Fooocus/modules/patch.py", line 465, in loader
result = original_loader(*args, **kwargs)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/serialization.py", line 998, in load
with _open_file_like(f, 'rb') as opened_file:
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/serialization.py", line 445, in _open_file_like
return _open_file(name_or_buffer, mode)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/serialization.py", line 426, in __init__
super().__init__(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: '$HOME/Fooocus/models/checkpoints/DreamShaperXL_Turbo_v2'
During handling of the above exception, another exception occurred:
=
Traceback (most recent call last):
File "$HOME/Fooocus/modules/async_worker.py", line 1462, in worker
handler(task)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/modules/async_worker.py", line 1153, in handler
tasks, use_expansion, loras, current_progress = process_prompt(async_task, async_task.prompt, async_task.negative_prompt,
File "$HOME/Fooocus/modules/async_worker.py", line 661, in process_prompt
pipeline.refresh_everything(refiner_model_name=async_task.refiner_model_name,
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/modules/default_pipeline.py", line 250, in refresh_everything
refresh_base_model(base_model_name, vae_name)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/modules/default_pipeline.py", line 74, in refresh_base_model
model_base = core.load_model(filename, vae_filename)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/env/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
File "$HOME/Fooocus/modules/core.py", line 147, in load_model
unet, clip, vae, vae_filename, clip_vision = load_checkpoint_guess_config(ckpt_filename, embedding_directory=path_embeddings,
File "$HOME/Fooocus/ldm_patched/modules/sd.py", line 431, in load_checkpoint_guess_config
sd = ldm_patched.modules.utils.load_torch_file(ckpt_path)
File "$HOME/Fooocus/ldm_patched/modules/utils.py", line 22, in load_torch_file
pl_sd = torch.load(ckpt, map_location=device, pickle_module=ldm_patched.modules.checkpoint_pickle)
File "$HOME/Fooocus/modules/patch.py", line 481, in loader
raise ValueError(exp)
ValueError: [Errno 2] No such file or directory: '$HOME/Fooocus/models/checkpoints/DreamShaperXL_Turbo_v2'
Total time: 2.53 seconds
### Additional information
_No response_
Strange, works for me without any issues, both Windows 11 and MacOS (case-sensitive volume). Are you using APFS or NTFS / encrypted or unencrypted / case sensitive or insensitive?
console log (2.5.2, upstream)
C:\AI\Fooocus\python_embeded\python.exe C:\AI\Fooocus\Fooocus\entry_with_update.py
Update failed.
'refs/remotes/origin/main_upstream'
Update succeeded.
[System ARGV] ['C:\\AI\\Fooocus\\Fooocus\\entry_with_update.py', '--disable-analytics', '--listen', '--always-download-new-model', '--language', 'en']
Python 3.10.9 (tags/v3.10.9:1dd9be6, Dec 6 2022, 20:01:21) [MSC v.1934 64 bit (AMD64)]
Fooocus version: 2.5.2
[Cleanup] Attempting to delete content of temp dir C:\Users\manue\AppData\Local\Temp\fooocus
[Cleanup] Cleanup successful
Total VRAM 10240 MB, total RAM 32693 MB
xformers version: 0.0.23
Set vram state to: NORMAL_VRAM
Always offload VRAM
Device: cuda:0 NVIDIA GeForce RTX 3080 : native
VAE dtype: torch.bfloat16
Using xformers cross attention
Refiner unloaded.
Running on local URL: http://0.0.0.0:7865
model_type EPS
UNet ADM Dimension 2816
To create a public link, set `share=True` in `launch()`.
Using xformers attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using xformers attention in VAE
extra {'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_l.logit_scale'}
left over keys: dict_keys(['cond_stage_model.clip_l.transformer.text_model.embeddings.position_ids'])
Base model loaded: C:\AI\Fooocus\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors
VAE loaded: None
Request to load LoRAs [('sd_xl_offset_example-lora_1.0.safetensors', 0.1)] for model [C:\AI\Fooocus\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors].
Loaded LoRA [C:\AI\Fooocus\Fooocus\models\loras\sd_xl_offset_example-lora_1.0.safetensors] for UNet [C:\AI\Fooocus\Fooocus\models\checkpoints\juggernautXL_v8Rundiffusion.safetensors] with 788 keys at weight 0.1.
Fooocus V2 Expansion: Vocab with 642 words.
C:\AI\Fooocus\python_embeded\lib\site-packages\torch\_utils.py:831: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage()
return self.fget.__get__(instance, owner)()
Fooocus Expansion engine loaded for cuda:0, use_fp16 = True.
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
[Fooocus Model Management] Moving model(s) has taken 0.67 seconds
Started worker with PID 13040
App started successful. Use the app with http://localhost:7865/ or 0.0.0.0:7865
Loaded preset: C:\AI\Fooocus\Fooocus\presets\turbo_custom.json
[Parameters] Adaptive CFG = 7
[Parameters] CLIP Skip = 2
[Parameters] Sharpness = 3
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] Seed = 5965757955318332010
[Parameters] CFG = 2
[Fooocus] Loading control models ...
[Parameters] Sampler = dpmpp_sde - karras
[Parameters] Steps = 8 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
model_type EPS
UNet ADM Dimension 2816
Using xformers attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using xformers attention in VAE
extra {'cond_stage_model.clip_l.text_projection', 'cond_stage_model.clip_l.logit_scale'}
left over keys: dict_keys(['conditioner.embedders.0.logit_scale', 'conditioner.embedders.0.text_projection'])
Base model loaded: C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors
VAE loaded: None
Request to load LoRAs [] for model [C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors].
Requested to load SDXLClipModel
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 0.73 seconds
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus] Encoding negative #1 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (1024, 1024)
Preparation time: 8.64 seconds
Using karras scheduler.
[Fooocus] Preparing task 1/1 ...
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.19 seconds
100%|██████████| 8/8 [00:06<00:00, 1.29it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.64 seconds
[Fooocus] Saving image 1/1 to system ...
Image generated with private log at: C:\AI\Fooocus\Fooocus\outputs\2024-07-29\log.html
Generating and saving time: 10.25 seconds
[Enhance] Skipping, preconditions aren't met
Processing time (total): 10.25 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 18.93 seconds
[Fooocus Model Management] Moving model(s) has taken 0.40 seconds
console log (2.6.2, my fork)
[Fooocus Model Management] Moving model(s) has taken 0.67 seconds
Started worker with PID 6888
App started successful. Use the app with http://localhost:7865/ or 0.0.0.0:7865
Loaded preset: C:\AI\Fooocus\Fooocus\presets\turbo_custom.json
Downloading: "https://huggingface.co/Lykon/dreamshaper-xl-v2-turbo/resolve/main/DreamShaperXL_Turbo_v2_1.safetensors" to C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors
100%|██████████| 6.46G/6.46G [01:16<00:00, 91.3MB/s]
[Parameters] Adaptive CFG = 7
[Parameters] CLIP Skip = 2
[Parameters] Sharpness = 3
[Parameters] ControlNet Softness = 0.25
[Parameters] ADM Scale = 1.5 : 0.8 : 0.3
[Parameters] Seed = 2207798739955052681
[Parameters] CFG = 2
[Fooocus] Loading control models ...
[Parameters] Sampler = dpmpp_sde - karras
[Parameters] Steps = 8 - 15
[Fooocus] Initializing ...
[Fooocus] Loading models ...
Refiner unloaded.
model_type EPS
UNet ADM Dimension 2816
Using xformers attention in VAE
Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
Using xformers attention in VAE
extra {'cond_stage_model.clip_l.logit_scale', 'cond_stage_model.clip_l.text_projection'}
left over keys: dict_keys(['conditioner.embedders.0.logit_scale', 'conditioner.embedders.0.text_projection'])
Base model loaded: C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors
VAE loaded: None
Request to load LoRAs [] for model [C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors].
Requested to load SDXLClipModel
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.75 seconds
[Fooocus] Processing prompts ...
[Fooocus] Encoding positive #1 ...
[Fooocus] Encoding negative #1 ...
[Parameters] Denoising Strength = 1.0
[Parameters] Initial Latent shape: Image Space (1024, 1024)
Preparation time: 11.29 seconds
Using karras scheduler.
[Fooocus] Preparing task 1/1 ...
[Sampler] refiner_swap_method = joint
[Sampler] sigma_min = 0.0291671771556139, sigma_max = 14.614643096923828
Requested to load SDXL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 2.25 seconds
100%|██████████| 8/8 [00:05<00:00, 1.36it/s]
Requested to load AutoencoderKL
Loading 1 new model
[Fooocus Model Management] Moving model(s) has taken 1.63 seconds
[Fooocus] Saving image 1/1 to system ...
[Cache] Calculating sha256 for C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors
[Cache] sha256 for C:\AI\Fooocus\Fooocus\models\checkpoints\DreamShaperXL_Turbo_v2_1.safetensors: 4496b36d48
Image generated with private log at: C:\AI\Fooocus\Fooocus\outputs\2024-07-29\log.html
Generating and saving time: 16.15 seconds
[Enhance] Skipping, preconditions aren't met
Processing time (total): 16.15 seconds
Requested to load SDXLClipModel
Requested to load GPT2LMHeadModel
Loading 2 new models
Total time: 27.46 seconds
[Fooocus Model Management] Moving model(s) has taken 0.59 seconds
Also works on Colab btw, so i assume it's an issue with your filesystem or the naming of the model file. Have you tried deleting the model and redownloading / 1:1 renaming as in your preset?
Very strange indeed. Nothing changed with my system as far as I can tell and I am having the same issue on MacOS and Windows 10.
Appreciate you validating that it is working everywhere else so that helps me determine the issue is on my system.