Mikubill / sd-webui-controlnet

WebUI extension for ControlNet
GNU General Public License v3.0
16.78k stars 1.93k forks source link

ControlNet not working after installing Linux Mint 21.3 #2973

Open JonnyNava367 opened 2 months ago

JonnyNava367 commented 2 months ago

I have a Dell G7 7588, with an i7-8750H, 20GB RAM, GTX 1060 Max-Q 6GB, a 256GB NVME, and a 1TB HDD. I'm using Linux Mint 21.3, and ControlNet worked fine before when I installed it on my laptop's HDD. But since I installed it on my laptop's NVME (which wiped my Windows install, since I no longer use it), it doesn't want to work, regardless of model or VAE. Here's the error when using ControlNet: /media/MyDrive/stable-diffusion-webui/modules/sd_samplers_common.py:68: RuntimeWarning: invalid value encountered in cast x_sample = x_sample.astype(np.uint8) This one happens when ControlNet is used.

/media/MyDrive/stable-diffusion-webui/modules/processing.py:1026: RuntimeWarning: invalid value encountered in cast x_sample = x_sample.astype(np.uint8) This other happens when finishing or interrupting the generation.

I also have these issues when I'm enabling ControlNet, enabling stuff like Low VRAM, or changing models:

2024-07-02 19:18:08,162 - ControlNet - INFO - [canny.threshold_a] Invalid value(0.5), using default value 100.
2024-07-02 19:18:08,162 - ControlNet - INFO - [canny.threshold_b] Invalid value(0.5), using default value 200.
2024-07-02 19:18:08,202 - ControlNet - INFO - [canny.threshold_a] Invalid value(0.5), using default value 100.
2024-07-02 19:18:08,202 - ControlNet - INFO - [canny.threshold_b] Invalid value(0.5), using default value 200.

And it produces a black image: Screenshot from 2024-07-02 19-20-34

I had this issue with A1111 v1.9.4, python 3.10.12, torch 2.1.2+cu121, xformers 0.0.23post1, gradio 3.41.2, torchvision 0.16.2+cu121, and cuda_11.5.r11.5/compiler.30672275_0 Even updating it to with A1111 v1.9.4, python 3.10.12, torch2.3.1+cu121, xformers 0.0.26.post1, gradio 3.41.2, torchvision 0.18.1, and cuda_11.5.r11.5/compiler.30672275_0, it still doesn't work. Here's my webui arguments: --xformers --deepdanbooru --disable-nan-check --medvram --api And here are the extensions I have installed: Screenshot from 2024-07-02 19-54-30 Even with a virgin Stable Diffusion install on my NVME, with just ControlNet installed, I get the same black image, this didn't happen before, and I don't know why. I even tried the ControlNet 1.0 version and I got the same issues. I'm using ControlNet v1.1.449.

huchenlei commented 2 months ago

Can you try disable fp8?

JonnyNava367 commented 2 months ago

How? Never heard of it

JonnyNava367 commented 2 months ago

This? Screenshot from 2024-07-06 17-40-55 It was already disabled.

JonnyNava367 commented 2 months ago

Here's a log abut what's hapening:

################################################################

################################################################
Repo already cloned, using it as install directory
################################################################

################################################################
Create and activate python venv
################################################################

################################################################
Launching launch.py...
################################################################
glibc version is 2.35
Check TCMalloc: libtcmalloc_minimal.so.4
libtcmalloc_minimal.so.4 is linked with libc.so,execute LD_PRELOAD=/lib/x86_64-linux-gnu/libtcmalloc_minimal.so.4
Python 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0]
Version: v1.9.4
Commit hash: feee37d75f1b168768014e4634dcb156ee649c05
Installing sd-webui-controlnet requirement: changing opencv-python version from 4.10.0.84 to 4.8.0
ControlNet init warning: Unable to install insightface automatically. Please try run `pip install insightface` manually.
Launching Web UI with arguments: --xformers --deepdanbooru --no-half-vae --disable-nan-check --medvram
ControlNet preprocessor location: /media/MyDrive/stable-diffusion-webui/extensions/sd-webui-controlnet/annotator/downloads
2024-07-06 17:50:58,003 - ControlNet - INFO - ControlNet v1.1.452
Loading weights [7f96a1a9ca] from /media/MyDrive/stable-diffusion-webui/models/Stable-diffusion/AnythingV5Ink_v5PrtRE.safetensors
2024-07-06 17:50:59,107 - ControlNet - INFO - ControlNet UI callback registered.
add tab
[([button, button, button, button, button], button, gallery), ([button, button, button, button, button], button, gallery)]
Creating model from config: /media/MyDrive/stable-diffusion-webui/configs/v1-inference.yaml
Running on local URL:  http://127.0.0.1:7860
/media/MyDrive/stable-diffusion-webui/venv/lib/python3.10/site-packages/huggingface_hub/file_download.py:1132: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.
  warnings.warn(
Loading VAE weights specified in settings: /media/MyDrive/stable-diffusion-webui/models/VAE/vae-ft-mse-840000-ema-pruned.safetensors
Applying attention optimization: xformers... done.
Model loaded in 3.2s (load weights from disk: 1.3s, create model: 0.5s, apply weights to model: 0.6s, load VAE: 0.1s, load textual inversion embeddings: 0.2s, calculate empty prompt: 0.4s).

To create a public link, set `share=True` in `launch()`.
Startup time: 15.9s (prepare environment: 4.5s, import torch: 3.1s, import gradio: 0.9s, setup paths: 0.9s, initialize shared: 0.1s, other imports: 0.4s, list SD models: 0.2s, load scripts: 2.3s, create ui: 1.0s, gradio launch: 2.4s).

2024-07-06 17:51:54,868 - ControlNet - INFO - unit_separate = False, style_align = False
2024-07-06 17:51:55,090 - ControlNet - WARNING - Unable to determine version for ControlNet model 'controlnet11Models_canny [b18e0966]'.
2024-07-06 17:51:55,090 - ControlNet - INFO - Loading model: controlnet11Models_canny [b18e0966]
2024-07-06 17:51:55,100 - ControlNet - INFO - Loaded state_dict from [/media/MyDrive/stable-diffusion-webui/extensions/sd-webui-controlnet/models/controlnet11Models_canny.safetensors]
2024-07-06 17:51:55,100 - ControlNet - INFO - controlnet_default_config
2024-07-06 17:51:57,574 - ControlNet - INFO - ControlNet model controlnet11Models_canny [b18e0966](ControlModelType.ControlNet) loaded.
2024-07-06 17:51:57,623 - ControlNet - INFO - Using preprocessor: canny
2024-07-06 17:51:57,623 - ControlNet - INFO - preprocessor resolution = 480
2024-07-06 17:51:57,671 - ControlNet - INFO - ControlNet Hooked - Time = 2.806868553161621
  0%|                                                                                                                                                                                                                  | 0/20 [00:00<?, ?it/s]/media/MyDrive/stable-diffusion-webui/modules/sd_samplers_common.py:68: RuntimeWarning: invalid value encountered in cast
  x_sample = x_sample.astype(np.uint8)
 40%|████████████████████████████████████████████████████████████████████████████████▊                                                                                                                         | 8/20 [00:12<00:18,  1.51s/it]
/media/MyDrive/stable-diffusion-webui/modules/processing.py:1026: RuntimeWarning: invalid value encountered in cast                                                                                            | 8/20 [00:10<00:15,  1.33s/it]
  x_sample = x_sample.astype(np.uint8)
Total progress:  40%|██████████████████████████████████████████████████████████████████████████▍                                                                                                               | 8/20 [00:11<00:17,  1.48s/it]
Total progress:  40%|██████████████████████████████████████████████████████████████████████████▍                                                                                                               | 8/20 [00:11<00:15,  1.33s/it]
JonnyNava367 commented 1 month ago

I installed Linux and Stable Diffusion again on my HDD, and it still doesn't work. I don't remember how I made ControlNet work on my first install.