continue writing the user guide
This commit is contained in:
parent
be16f33151
commit
b508da747e
|
@ -87,15 +87,27 @@ produce the same image.
|
||||||
|
|
||||||
#### CFG parameter
|
#### CFG parameter
|
||||||
|
|
||||||
Classifier free guidance.
|
Classifier free guidance. How strictly the model should follow the prompt. Anything from 5 to 15 usually works. More is
|
||||||
|
not always better, setting this too high can result in noisy, solarized images.
|
||||||
|
|
||||||
|
Roughly:
|
||||||
|
|
||||||
|
- 2-6 allows the AI to be creative
|
||||||
|
- 7-11 treats the prompt as a suggestion
|
||||||
|
- 12-15 strongly encourages the AI to follow the prompt
|
||||||
|
- 16-20 follows the prompt whether it makes sense or not
|
||||||
|
|
||||||
#### Steps parameter
|
#### Steps parameter
|
||||||
|
|
||||||
The number of scheduler steps to run.
|
The number of scheduler steps to run. Using more steps often results in an image with more details, but also takes
|
||||||
|
longer to run. The Euler Ancestral scheduler can usually produce decent results in 30-45 steps, while some of the
|
||||||
|
others need 80-100 or more. Inpainting may need more steps, up to 120 or 150 in some cases.
|
||||||
|
|
||||||
#### Seed parameter
|
#### Seed parameter
|
||||||
|
|
||||||
The random seed.
|
The seed value used for the random number generators.
|
||||||
|
|
||||||
|
Using -1 will generate a new seed on the server for each image.
|
||||||
|
|
||||||
#### Prompt parameter
|
#### Prompt parameter
|
||||||
|
|
||||||
|
@ -111,6 +123,8 @@ Resize the output image before returning it to the client.
|
||||||
|
|
||||||
Enabling this will run Real ESRGAN and requires an upscaling model.
|
Enabling this will run Real ESRGAN and requires an upscaling model.
|
||||||
|
|
||||||
|
Check out [the Real ESRGAN Github](https://github.com/xinntao/Real-ESRGAN) for more details.
|
||||||
|
|
||||||
#### Scale parameter
|
#### Scale parameter
|
||||||
|
|
||||||
The output scale for Real ESRGAN.
|
The output scale for Real ESRGAN.
|
||||||
|
@ -122,12 +136,16 @@ output. Lanczos interpolation is used when the outscale is greater than the scal
|
||||||
|
|
||||||
#### Denoise parameter
|
#### Denoise parameter
|
||||||
|
|
||||||
|
The amount of denoising to apply when using the RealESR x4 v4 model. Can be used to avoid over-smoothing the results.
|
||||||
|
|
||||||
#### Face correction and strength
|
#### Face correction and strength
|
||||||
|
|
||||||
Run face correction the the output image before returning it to the client.
|
Run face correction the the output image before returning it to the client.
|
||||||
|
|
||||||
Enabling this will run GFPGAN and requires a correction model.
|
Enabling this will run GFPGAN and requires a correction model.
|
||||||
|
|
||||||
|
Check out [the GFPGAN Github](https://github.com/TencentARC/GFPGAN) for more details.
|
||||||
|
|
||||||
### Scheduler comparison
|
### Scheduler comparison
|
||||||
|
|
||||||
https://huggingface.co/docs/diffusers/main/en/using-diffusers/schedulers#compare-schedulers
|
https://huggingface.co/docs/diffusers/main/en/using-diffusers/schedulers#compare-schedulers
|
||||||
|
|
Loading…
Reference in New Issue