Developing custom scripts - linwownil/stable-diffusion-webui GitHub Wiki
The Script class definition can be found in modules/scripts.py
. To create your own custom script, create a python script that implements the class and drop it into the scripts
folder, using the below example or other scripts already in the folder as a guide.
The Script class has four primary methods, described in further detail below with a simple example script that rotates and/or flips generated images.
import modules.scripts as scripts
import gradio as gr
import os
from modules import images
from modules.processing import process_images, Processed
from modules.processing import Processed
from modules.shared import opts, cmd_opts, state
class Script(scripts.Script):
# The title of the script. This is what will be displayed in the dropdown menu.
def title(self):
return "Flip/Rotate Output"
# Determines when the script should be shown in the dropdown menu via the
# returned value. As an example:
# is_img2img is True if the current tab is img2img, and False if it is txt2img.
# Thus, return is_img2img to only show the script on the img2img tab.
def show(self, is_img2img):
return is_img2img
# How the script's is displayed in the UI. See https://gradio.app/docs/#components
# for the different UI components you can use and how to create them.
# Most UI components can return a value, such as a boolean for a checkbox.
# The returned values are passed to the run method as parameters.
def ui(self, is_img2img):
angle = gr.Slider(minimum=0.0, maximum=360.0, step=1, value=0,
label="Angle")
hflip = gr.Checkbox(False, label="Horizontal flip")
vflip = gr.Checkbox(False, label="Vertical flip")
overwrite = gr.Checkbox(False, label="Overwrite existing files")
return [angle, hflip, vflip, overwrite]
# This is where the additional processing is implemented. The parameters include
# self, the model object "p" (a StableDiffusionProcessing class, see
# processing.py), and the parameters returned by the ui method.
# Custom functions can be defined here, and additional libraries can be imported
# to be used in processing. The return value should be a Processed object, which is
# what is returned by the process_images method.
def run(self, p, angle, hflip, vflip, overwrite):
# function which takes an image from the Processed object,
# and the angle and two booleans indicating horizontal and
# vertical flips from the UI, then returns the
# image rotated and flipped accordingly
def rotate_and_flip(im, angle, hflip, vflip):
from PIL import Image
raf = im
if angle != 0:
raf = raf.rotate(angle, expand=True)
if hflip:
raf = raf.transpose(Image.FLIP_LEFT_RIGHT)
if vflip:
raf = raf.transpose(Image.FLIP_TOP_BOTTOM)
return raf
# If overwrite is false, append the rotation information to the filename
# using the "basename" parameter and save it in the same directory.
# If overwrite is true, stop the model from saving its outputs and
# save the rotated and flipped images instead.
basename = ""
if(not overwrite):
if angle != 0:
basename += "rotated_" + str(angle)
if hflip:
basename += "_hflip"
if vflip:
basename += "_vflip"
else:
p.do_not_save_samples = True
proc = process_images(p)
# rotate and flip each image in the processed images
# use the save_images method from images.py to save
# them.
for i in range(len(proc.images)):
proc.images[i] = rotate_and_flip(proc.images[i], angle, hflip, vflip)
images.save_image(proc.images[i], p.outpath_samples, basename,
proc.seed + i, proc.prompt, opts.samples_format, info= proc.info, p=p)
return proc