This is a program that redraws pictures.
The program will read all files in
img_dir_path
inconfig.ini
and then perform style conversion.
-
Clone Repository
git clone https://github.com/danielchan-25/redraw_img.git
-
Install dependencies
pip install -r requirements.txt
python main.py
See config.ini
Argument | Type | Description | Example |
---|---|---|---|
url |
str | STable Diffusion API Url address | http://127.0.0.1:7860 |
img_dir_path |
str | Image Dir Path | E:\temp |
sd_model_checkpoint |
str | Model Name | v1-5-pruned-emaonly |
prompt |
str | prompt | |
negative_prompt |
str | negative prompt | |
steps |
int | Step count | 20 |
width |
int | Image width | 512 |
height |
int | Image height | 512 |
seed |
int | seed | -1 |
samples |
int | samples | 1 |
batch_size |
int | batch size | 1 |
sampler_name |
str | sampler name | Euler a |
cfg_scale |
int | Prompt word relevance | 7 |
denoising_strength |
float | denoising strength | 0.7 |
restore_faces |
bool | restore faces | True |
save_images |
bool | save images to local | True |