search

AUTOMATIC1111 / stable-diffusion-webui

Stable Diffusion web UI
GNU Affero General Public License v3.0
139.41k stars 26.43k forks source link

[Bug]: Just resize (latent upscale) causes "size of tensors must match..." errors #10157

Open PaperOrb opened 1 year ago

PaperOrb commented 1 year ago

Is there an existing issue for this?

What happened?

I keep getting "RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 64 but got size 112 for tensor number 1 in the list" no matter what resolution I try. Here's res's I've tried so far:

  1. 512x512
  2. 960x896
  3. <insert dimension that's a multiple of 64px here>
  4. and other random reses

Steps to reproduce the problem

  1. Go to img2img
  2. tick the "Just resize (latent upscale)" button
  3. try generating with the resolutions listed

What should have happened?

The image should generate without this error appearing.

Commit where the problem happens

5ab7f213

What platforms do you use to access the UI ?

Linux

What browsers do you use to access the UI ?

Mozilla Firefox

Command Line Arguments

export COMMANDLINE_ARGS="--opt-sdp-attention --theme dark --no-half --medvram --precision full"

List of extensions

  1. a1111-sd-webui-tagcomplete
  2. sd-dynamic-prompts

Console logs

cd ~/Desktop/stable-diffusion-webui && ./webui.sh

################################################################
Install script for stable-diffusion + Web UI
Tested on Debian 11 (Bullseye)
################################################################

################################################################
Running on user user
################################################################

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

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

################################################################
Launching launch.py...
################################################################
Using TCMalloc: libtcmalloc.so.4
Python 3.10.6 (main, Mar 10 2023, 10:55:28) [GCC 11.3.0]
Commit hash: 5ab7f213bec2f816f9c5644becb32eb72c8ffb89
Installing requirements
Installing sd-dynamic-prompts requirements.txt

Launching Web UI with arguments: --opt-sdp-attention --theme dark --no-half --medvram --precision full
No module 'xformers'. Proceeding without it.
Loading weights [9140e020d1] from /home/pc/Desktop/stable-diffusion-webui/models/Stable-diffusion/mixProV4-inpainting.inpainting.safetensors
Creating model from config: /home/pc/Desktop/stable-diffusion-webui/configs/v1-inpainting-inference.yaml
LatentInpaintDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.54 M params.
Loading VAE weights specified in settings: /home/pc/Desktop/stable-diffusion-webui/models/VAE/clearvae_main.safetensors
Applying scaled dot product cross attention optimization.
Textual inversion embeddings loaded(0): 
Model loaded in 2.5s (load weights from disk: 0.3s, create model: 0.4s, apply weights to model: 1.5s, load VAE: 0.3s).
Running on local URL:  http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
Startup time: 6.1s (import torch: 0.9s, import gradio: 0.7s, import ldm: 0.3s, other imports: 0.6s, load scripts: 0.4s, load SD checkpoint: 2.6s, create ui: 0.4s, scripts app_started_callback: 0.1s).
100%|████████████████████████████████████████████████████████████████████████████████████████████████████| 13/13 [00:19<00:00,  1.52s/it]
Total progress: 100%|████████████████████████████████████████████████████████████████████████████████████| 13/13 [00:25<00:00,  1.92s/it]
Error completing request█████████████████████████████████████████████████████████████████████████████████| 13/13 [00:25<00:00,  1.33s/it]
Arguments: ('task(6y12m9xtlutks2w)', 0, 'high quality, goku', 'worst quality, low quality, monochrome, greyscale, (muted color, pale color:1.3)', [], <PIL.Image.Image image mode=RGBA size=800x900 at 0x7FD2CFB575B0>, None, None, None, None, None, None, 20, 3, 4, 0, 1, False, False, 1, 1, 7, 1.5, 0.6, -1.0, -1.0, 0, 0, 0, False, 0, 512, 512, 0.65, 3, 0, 32, 0, '', '', '', [], 0, True, False, 1, False, False, False, 1.1, 1.5, 100, 0.7, False, False, True, False, False, 0, 'Gustavosta/MagicPrompt-Stable-Diffusion', '', '<ul>\n<li><code>CFG Scale</code> should be 2 or lower.</li>\n</ul>\n', True, True, '', '', True, 50, True, 1, 0, False, 4, 0.5, 'Linear', 'None', '<p style="margin-bottom:0.75em">Recommended settings: Sampling Steps: 80-100, Sampler: Euler a, Denoising strength: 0.8</p>', 128, 8, ['left', 'right', 'up', 'down'], 1, 0.05, 128, 4, 0, ['left', 'right', 'up', 'down'], False, False, 'positive', 'comma', 0, False, False, '', '<p style="margin-bottom:0.75em">Will upscale the image by the selected scale factor; use width and height sliders to set tile size</p>', 64, 0, 2, 1, '', [], 0, '', [], 0, '', [], True, False, False, False, 0) {}
Traceback (most recent call last):
  File "/home/pc/Desktop/stable-diffusion-webui/modules/call_queue.py", line 57, in f
    res = list(func(*args, **kwargs))
  File "/home/pc/Desktop/stable-diffusion-webui/modules/call_queue.py", line 37, in f
    res = func(*args, **kwargs)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/img2img.py", line 181, in img2img
    processed = process_images(p)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/processing.py", line 515, in process_images
    res = process_images_inner(p)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/processing.py", line 604, in process_images_inner
    p.init(p.all_prompts, p.all_seeds, p.all_subseeds)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/processing.py", line 1106, in init
    self.image_conditioning = self.img2img_image_conditioning(image, self.init_latent, image_mask)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/processing.py", line 260, in img2img_image_conditioning
    return self.inpainting_image_conditioning(source_image, latent_image, image_mask=image_mask)
  File "/home/pc/Desktop/stable-diffusion-webui/modules/processing.py", line 243, in inpainting_image_conditioning
    image_conditioning = torch.cat([conditioning_mask, conditioning_image], dim=1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 64 but got size 112 for tensor number 1 in the list.

Additional information

No response

allo- commented 11 months ago

I have the same problem, but only on one of two machines. Both use python3.10 and are at commit 5ef669de080814067961f28357256e8fe27544f4.

allo- commented 11 months ago 
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 192 but got size 96 for tensor number 1 in the list.

For 512x768 upscaling to 1024xx1536, so using multiple of 64.