Home Artists Posts Import Register

Downloads

Content

Patreon exclusive posts index

Join discord and tell me your discord username to get a special rank : SECourses Discord

20 March 2024 Update:

  • The installer files and Gradio app updated to work blazing fast with latest libraries
  • Free Kaggle notebook is made and working amazing
  • The Kaggle notebook is included in shopify_v3.zip file 

Requirements:

Installation:

Download the attached shopify_v3.zip and extract into a folder where you want to install (no SPACES or non-English characters super important, e.g.: bg_replace) where you want to install.

Double click and start windows_install .bat file. This will install everything automatically for you

After that we will use start_background_replacement_app .bat file to start the Shopify Background replacement app

When you use the start_background_replacement_app .bat first time running it will download necessary model files into your cache folder

e.g. : C:\Users\King\.cache\huggingface\hub

If you want to change your cache folder watch this tutorial i have explained : https://youtu.be/rjXsJ24kQQg

The app started url will be by default http://127.0.0.1:7860/ and it will automatically open in browser

1024x1024 pixel resolution of images are the best for this APP but looks like supporting all resolutions

Follow the instructions written on the Gradio

Let me know also if you find better settings

27 September 2023 Major Update

  • Added so many new options
  • Also it will now auto save all images inside outputs folder

Here few examples

When you look at the CMD you will see how it is prompting


Comments

Anonymous

Hello.🤝 Startup problem - Auto Windows Installer For SDXL Background Replacement for Product Images A matching Triton is not available, some optimizations will not be enabled. Error caught was: No module named 'triton' Initializing captioner... Initializing upscaler... Initializing segmenter... Initializing depth estimator... Initializing depth ControlNet... Initializing autoencoder... Initializing SDXL pipeline... Loading pipeline components...: 100%|████████████████████████████████████████████████████| 7/7 [00:08<00:00, 1.26s/it] Traceback (most recent call last): File "D:\pimages\background-replacement\app.py", line 150, in num_inference_steps = gr.Slider(precision=0,minimum=1, maximum=300,value=30, default=30,step=1, label="Num Inference Steps") File "D:\pimages\background-replacement\venv\lib\site-packages\gradio\component_meta.py", line 146, in wrapper return fn(self, **kwargs) TypeError: Slider.init() got an unexpected keyword argument 'precision' To continue, press any key. . .

Anonymous

Traceback (most recent call last): File "D:\pimages\background-replacement\app.py", line 150, in num_inference_steps = gr.Slider(precision=0,minimum=1, maximum=300,value=30, default=30,step=1, label="Num Inference Steps") File "D:\pimages\background-replacement\venv\lib\site-packages\gradio\component_meta.py", line 146, in wrapper return fn(self, **kwargs) TypeError: Slider.__init__() got an unexpected keyword argument 'precision' To continue, press any key. . . help me solve the problem Thank you

Anonymous

Hi thanks for replying, I am using python 3,11 and just re-installed it on disc C and get this message: Traceback (most recent call last): File "C:\backgroundrpl\background-replacement\app.py", line 2, in import gradio as gr ModuleNotFoundError: No module named 'gradio' Press any key to continue . . .

Furkan Gözükara

hi. sadly 3.11 will not work. please install 3.10.11 and make it default as shown here : https://youtu.be/-NjNy7afOQ0 also install c++ tools and ffmpeg too

Jesse

This is very cool. Are we able to change the model to a different SDXL model?

Andrea Rotondo

good morning where can i change the SDXL MODEL?

Furkan Gözükara

in the pipeline.py. you need to find hugging face repo that includes diffusers files and then you can change : pipe = StableDiffusionXLControlNetPipeline.from_pretrained( i dont know if single safetensors would work but you can try with giving full path. i will try in a moment