[Stable Diffusion] Prompt Sharing and Learning Thread

[me3]

Imma start on Lara Croft LORA: https://f95zone.to/threads/loras-for-wildeers-lara-croft-development-thread.173873/

RE this, i had to get the training stuff working again so i grabbed some of the images for a quick test (quick as in prep, not actual AI training).
Seeing some issues with the captions and potentially my choice of images used, but it seems to be working. One (or more) images clearly has some wet/oily skin that's not captioned out, but hey maybe it's sweaty in the jungle, not that uncommon


(image got resized despite me cutting it up, i fully understand the byte limit. but less of a fan of the pixel limit, anyway...It's only by a few hundred px)

 

[Sharinel]

RE this, i had to get the training stuff working again so i grabbed some of the images for a quick test (quick as in prep, not actual AI training).
Seeing some issues with the captions and potentially my choice of images used, but it seems to be working. One (or more) images clearly has some wet/oily skin that's not captioned out, but hey maybe it's sweaty in the jungle, not that uncommon
View attachment 2915252

(image got resized despite me cutting it up, i fully understand the byte limit. but less of a fan of the pixel limit, anyway...It's only by a few hundred px)

Is this Wildeer's Lara? Must say the face doesn't look like it in any of the pics to me?

 

[me3]

Is this Wildeer's Lara? Must say the face doesn't look like it in any of the pics to me?

can't say, all images are from that thread but i have no idea who made them. images aren't consistent in features and some of them have watermarks suggesting it's different creators. If you go though all the images you clearly see there's differences both in body and face.

 

[Sepheyer]

can't say, all images are from that thread but i have no idea who made them. images aren't consistent in features and some of them have watermarks suggesting it's different creators. If you go though all the images you clearly see there's differences both in body and face.

That's actually very-very good.

 

[Sepheyer]

RE this, i had to get the training stuff working again so i grabbed some of the images for a quick test (quick as in prep, not actual AI training).
Seeing some issues with the captions and potentially my choice of images used, but it seems to be working. One (or more) images clearly has some wet/oily skin that's not captioned out, but hey maybe it's sweaty in the jungle, not that uncommon
View attachment 2915252

(image got resized despite me cutting it up, i fully understand the byte limit. but less of a fan of the pixel limit, anyway...It's only by a few hundred px)

Can you please post in the Lara's LORA thread:
- which images you used (just a screenshot of your training image folder)
- what your tags look like
- what model was used for training (ideally if you could post your Kohya settings)
- one of the files that contains a prompt
- and how you got the green leotard to come up

 

[me3]

Can you please post in the Lara's LORA thread:
- which images you used (just a screenshot of your training image folder)
- what your tags look like
- what model was used for training (ideally if you could post your Kohya settings)
- one of the files that contains a prompt
- and how you got the green leotard to come up

Don't have time right atm, but will try later. As quick run down:
- <tbd>
- horrible, it was just blip without beams as that is still broken for me, yay
- settings is exactly the same as i used in the lora i posted a while ago, constant, 0.0001 lr, 128/64 network, adamw 8bit, cyber_3.3, can't remember there being much else settings atm
- prompt for generation was basically just "woman dressed for jungle exploration", AI clearly think you don't need too much clothing, probably due to heat and you see she looks drenched in sweat
- "nothing", coloring the AI picked up itself, it really likes using it for clothing, rest is more then likely to not having captioned anything about the clothing. captions are so basic they mentioned next to nothing besides, woman, <optional pose>, <optional location>. Lack of beams with blip is horrible.

 

[Sepheyer]

Don't have time right atm, but will try later. As quick run down:
- <tbd>
- horrible, it was just blip without beams as that is still broken for me, yay
- settings is exactly the same as i used in the lora i posted a while ago, constant, 0.0001 lr, 128/64 network, adamw 8bit, cyber_3.3, can't remember there being much else settings atm
- prompt for generation was basically just "woman dressed for jungle exploration", AI clearly think you don't need too much clothing, probably due to heat and you see she looks drenched in sweat
- "nothing", coloring the AI picked up itself, it really likes using it for clothing, rest is more then likely to not having captioned anything about the clothing. captions are so basic they mentioned next to nothing besides, woman, <optional pose>, <optional location>. Lack of beams with blip is horrible.

Great. My LORA is still training, around 20 hours left out of 50. Gonna suck major dick if I picked bad settings.

Meanwhile, I did a few renders without the LORA asking for Lara Croft / Tomb Raider. All of them are pretty far from Wildeer's, so we'll know right away is the LORA comes out aight.

The leotard is what makes me uneasy. I hope the LORA gets it out of the box, but I know I put zero effort there. Now debating if the second run should have tags such as: WildeerLara nude / WildeerLara in leotard and shorts / WildeerLara in leotard. Gonna be an excuse to learn the Kohya's "continue training" feature with a smaller subset on top of an already baked LORA.

Spoiler: More

You don't have permission to view the spoiler content. Log in or register now.

 

[rogue_69]

Just a link for a post I made about roop and the code edit you need to turn of the nsfw filter.
https://f95zone.to/threads/stable-diffusion-prompt-sharing-and-learning-thread.146036/post-11443616

A few examples:

[TR]
[TD]London Keys[/TD]
[TD]Alicia Vikander[/TD]
[TD]Jennifer Lawrence[/TD]
[TD]Gabrielle Union[/TD]
[/TR]
[TR]
[TD] View attachment 2910883 [/TD]
[TD] View attachment 2910884 [/TD]
[TD] View attachment 2910885 [/TD]
[TD] View attachment 2910886 [/TD]
[/TR]

I used roop with img2img, also known as photobashing. The face is only copy pasted over the original with this method so the shape is not changed. However when you use roop with txt2img roop is included in the generative process so the face shape is changed. I guess with a higher denoising setting you could possibly get a better result with img2img, I only did this very fast as an example.

I just got done messing around with Roop, but I'll have to check out turning off the NSFW filter. My end goal is doing animations, which I've gotten good results already, just not as good as I want. I think Roop might really help. I'm happy with the renders of my character's bodies, clothing, the background, etc. It's just touching up the face that I use Stable Diffusion for. I'm thinking Roop really might help eliminate more of the flicker you get with AI images and animation.

 

[Mr-Fox]

I just got done messing around with Roop, but I'll have to check out turning off the NSFW filter. My end goal is doing animations, which I've gotten good results already, just not as good as I want. I think Roop might really help. I'm happy with the renders of my character's bodies, clothing, the background, etc. It's just touching up the face that I use Stable Diffusion for. I'm thinking Roop really might help eliminate more of the flicker you get with AI images and animation.

It's very easy to edit the code to turn off the nsfw filter.
Go to: stable-diffusion-webui\extensions\sd-webui-roop\scripts\

cimage.py

either right mouse click and choose edit or open an empty notepad doc and drag/drop the file on to notepad.
Erase all code in cimage.py
Copy/paste this code instead (minus the first and last quotation marks""):

"import tempfile
def convert_to_sd(img):
return [False, tempfile.NamedTemporaryFile(delete=False, suffix=".png")]"

then save, done!

The nsfw filter is really janky and might give false positives as well as missing others, so it's not only to bypass the filter that we edit the code, it will make roop function much better in all scenarios.

 

[Mr-Fox]

Don't have time right atm, but will try later. As quick run down:
- <tbd>
- horrible, it was just blip without beams as that is still broken for me, yay
- settings is exactly the same as i used in the lora i posted a while ago, constant, 0.0001 lr, 128/64 network, adamw 8bit, cyber_3.3, can't remember there being much else settings atm
- prompt for generation was basically just "woman dressed for jungle exploration", AI clearly think you don't need too much clothing, probably due to heat and you see she looks drenched in sweat
- "nothing", coloring the AI picked up itself, it really likes using it for clothing, rest is more then likely to not having captioned anything about the clothing. captions are so basic they mentioned next to nothing besides, woman, <optional pose>, <optional location>. Lack of beams with blip is horrible.

You can always use the Interrogate functions in A1111 img2img, it's much slower ofc because you have to do one image at a time. When I start on a fresh prompt, sometimes I use an image with both clip and DeepBooru and then take the best from both, then I adjust and add my own stuff etc. Even though it's a big pain in da ass it's well worth taking the time to get really good captions when training a Lora, because good captions is very important to get a good result. According to what I have read, the training is more sensitive and less forgiving to bad captions than the quality of the images used for the training. Also if you are not careful a tag that is present often enough because of an item in the images that you are describing is consistent in many images, this can create unintentional trigger Words. I wasn't aware when I trained my Kendra Lora so "headband" became one of the trigger words.
In some images there was a fence in the background and it became trained into the Lora, so sometimes I need to add it to negative prompt with a strong weight to avoid it in the generated images. In order to get good sample result in the training you have to describe what is in the image and there are only so many ways you can word it, so then it can't be helped. You just need to do your best and cross your fingers. Sometimes..

 

[devilkkw]

have you updated a1111 to 1.6.0?
after update i'm unable to load any model, keep OOM error. so i switched back to 1.5.2 but i've modified sampler to keep dpm 3.
anyone had some issue on 1.6.0?
i not tested clean install, maybe it solves problem, but for now i keep 1.5.2.

 

[me3]

have you updated a1111 to 1.6.0?
after update i'm unable to load any model, keep OOM error. so i switched back to 1.5.2 but i've modified sampler to keep dpm 3.
anyone had some issue on 1.6.0?
i not tested clean install, maybe it solves problem, but for now i keep 1.5.2.

delete the venv folder, there's a bunch of conflicting dependencies etc so a "fresh" install of that fixes the issues in most cases, it's what i had to do for it, most of the installs will just use your cached files anyway so there isn't that much new downloading.

 

[pazhentaigame]

anyone Know how to do a step separate generate
sometime the img while it generating and the outcome is just too much difference
I just want that half way generation preview version
but interruption it will result in unfinished generate img instead
I don't even know it possible to get that version

 

[Mr-Fox]

anyone Know how to do a step separate generate
sometime the img while it generating and the outcome is just too much difference
I just want that half way generation preview version
but interruption it will result in unfinished generate img instead
I don't even know it possible to get that version

Use fewer steps. In the live preview settings you can decide how often the preview is updated, meaning how many steps in between updates. Lets say it is set to 10 steps and you are using 20 sample steps for generating images, then the preview that you like is somewhere between 10-20 steps. I would set the seed to static by copying the seed from that image that you liked the first preview. Then use xyz plot to test how many steps you should use to get it. Go to xyz plot in the scripts menu and select steps for x-axis.
Set it to 10-20 [11], now it will generate one image for each amount of steps (10,11,12..) in other words 11 images.
If you want fewer increments and thus fewer images simply set it to 10-20 [6] it will use the increments of 2 meaning 10,12,14 etc and generate 6 images.
Ok now that you found the amount of steps you can use xyz plot to try out the cfg scale in similair way, I would recommend to have your prompt mostly finalized before testing cfg scale. In xyz plot select cfg and set it to 4-12 [5] for increments of 2 and [9] for increments of 1.

 

[devilkkw]

delete the venv folder, there's a bunch of conflicting dependencies etc so a "fresh" install of that fixes the issues in most cases, it's what i had to do for it, most of the installs will just use your cached files anyway so there isn't that much new downloading.

made a full clean install, worked. but loading 9gb model keep getting OOM, in 1.5.2 i don't get any error.
It's strange, need inspecting more, i think some setting's need to be adjust. And wait next update maybe something wrong.
The 1.6.0 have some implementation for better working with XL model, memory management is different. But why loading 1.5 model upper to 4Gb get OOM now is a mystery.

 

[me3]

made a full clean install, worked. but loading 9gb model keep getting OOM, in 1.5.2 i don't get any error.
It's strange, need inspecting more, i think some setting's need to be adjust. And wait next update maybe something wrong.
The 1.6.0 have some implementation for better working with XL model, memory management is different. But why loading 1.5 model upper to 4Gb get OOM now is a mystery.

1.6.0 moved alot of the launcher settings from the launched to the UI, low/med vram is still in the launcher but xformers/sdp and other optimising settings are in the UI itself and some/all of those settings no longer work unless set in the UI.
They even moved a thing like the face restore option, (not found a way to restore it to its old place yet) which gets really annoying in the case where you have to keep turn it it on and off

 

[devilkkw]

thank you. try anything in setting's, unable to load big model. i'll stay on 1.5.2.

 

[Mr-Fox]

1.6.0 moved alot of the launcher settings from the launched to the UI, low/med vram is still in the launcher but xformers/sdp and other optimising settings are in the UI itself and some/all of those settings no longer work unless set in the UI.
They even moved a thing like the face restore option, (not found a way to restore it to its old place yet) which gets really annoying in the case where you have to keep turn it it on and off

You can add a face restore check box in the ui settings/quick settings, select face_restoration,
I also have face_restoration_model wich is convenient for switching between CodeFormer and GFPGAN.

 

[Sepheyer]

Edit.
LORA's here: link.
---
LORA coming soon, just need to test more and figure out how to upload.

Spoiler: More

You don't have permission to view the spoiler content. Log in or register now.

 

[me3]

Regarding this

So the face is still slightly off, couple of tattoos have gone missing and.... - "you're waiting? what you mean, waiting for what?"

Spoiler: prompt

You don't have permission to view the spoiler content. Log in or register now.