[Bug]: Saving individual images from X/Y/Z with batch count > 1

[MIC132]

Is there an existing issue for this?

• [X] I have searched the existing issues and checked the recent builds/commits

What happened?

I'm not sure if this is a bug as such, but it certainly seems wrong/unintended.

When generating a grid with batch count higher than 1, and checking "include individual images" the results include the grid, and the individual image batches, but not the individual images from those batches.

So for example when you do a X/Y with two "fields" and batch count of 4, the grid has 8 images in it, and you get the 2 separate 4-image batch images, but you don't get the 8 separate images.

Perhaps it should be a separate checkbox, but it's rather annoying. As is you need to re-generate a batch of a specific field in the grid (without X/Y) to get the individual images.

Steps to reproduce the problem

Generate a X/Y plot with batch count higher than 1 and "include individual images" checked.

What should have happened?

You should get the grid, the separate batches and the separate images from those batches.

Commit where the problem happens

ea9bd9fc7409109adcd61b897abc2c8881161256

What platforms do you use to access the UI ?

Windows

What browsers do you use to access the UI ?

Google Chrome

Command Line Arguments

--medvram --no-half-vae

List of extensions

a1111-sd-webui-tagcomplete

Console logs

venv "E:\stable-diffusion-webui\venv\Scripts\Python.exe"
Python 3.10.5 (tags/v3.10.5:f377153, Jun  6 2022, 16:14:13) [MSC v.1929 64 bit (AMD64)]
Commit hash: ea9bd9fc7409109adcd61b897abc2c8881161256
Installing requirements for Web UI
Launching Web UI with arguments: --medvram --no-half-vae
No module 'xformers'. Proceeding without it.
Loading weights [23362b0c63] from E:\stable-diffusion-webui\models\Stable-diffusion\ProtoThing200.ckpt
Creating model from config: E:\stable-diffusion-webui\configs\v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Loading VAE weights specified in settings: E:\stable-diffusion-webui\models\Stable-diffusion\AnythingV3.vae.pt
Applying cross attention optimization (Doggettx).
Model loaded in 79.5s (load weights from disk: 67.2s, create model: 1.8s, apply weights to model: 1.1s, apply half(): 1.1s, load VAE: 7.5s, load textual inversion embeddings: 0.8s).
Running on local URL:  http://127.0.0.1:7860

To create a public link, set `share=True` in `launch()`.
X/Y/Z plot will create 8 images on 1 2x1 grid; 4 images per cell. (Total steps to process: 200)
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:26<00:00,  1.06s/it]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.36it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.34it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.34it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.34it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.34it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.34it/s]
100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25/25 [00:18<00:00,  1.33it/s]
Total progress: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 200/200 [03:05<00:00,  1.08it/s]

Additional information

[EllangoK]

Not really a bug, I did notice it when updating the xy to xyz plot, and discussed it. The problem is that it often scales up very quickly, and having tons of images stored as b64 in the browser does reduce performance unfortunately. You are only using 1 axis with 2 settings currently and already have 8 images. It easily becomes hundreds of images using a full X/Y/Z plot which a lot of people often do. I guess it could be once again another checkbox, but the ui is cluttered already so idk.

Important though the images are there still, you shouldn't need to regenerate them. The sub batch images would be in your txt2img folder or wherever if you have Always save all generated images set in the Settings.

Also it is no longer called "include individual images", I don't know how you got that and are on the latest commit, but it should say "Include Sub Images". Regardless I guess it should be renamed to "Include Grid Cells", to be actually accurate.

[MIC132]

Also it is no longer called "include individual images", I don't know how you got that and are on the latest commit, but it should say "Include Sub Images". Regardless I guess it should be renamed to "Include Grid Cells", to be actually accurate.

I was quoting from memory, to be fair.

Important though the images are there still, you shouldn't need to regenerate them. The sub batch images would be in your txt2img folder or wherever if you have Always save all generated images set in the Settings.

I tend to have auto-saving disabled and only save results using the SAVE button when I like what I got. Some middle ground would be nice, otherwise I need to turn of always save specifically for batch grids.

[Corgano]

Came here for this exact issue. When you generate a batch of images, it shows you a grid, and then the individual images. This would be fantastic behavior for xyz grids, where you can check a box, and it will then go through and generate the grid, but at the end it will show the grid image, as well as the individual images it generated along the way. It doesn't need to save these pernamently, but then fi there were a few good generations in a xyz grid you could save them and pop them into png Info later on to use that specific setting.

[bluelovers]

need this