Stable-Diffusion-NCNN

Stable-Diffusion-NCNN copied to clipboard

int* token_ptr = token_mat; float* multiplier_ptr = multiplier_mat;

henbucuoshanghai

...

JohnClaw

...

JohnClaw

[( *u_ptr ) = ( *u_ptr ) + 7 * ( ( *c_ptr ) - ( *u_ptr ) );]

1

[( *u_ptr ) = ( *u_ptr ) + 7 * ( ( *c_ptr ) - ( *u_ptr ) );] https://github.com/fengwang/Stable-Diffusion-NCNN/blob/e3ba9d807c0c40197c8b13be2d94c690b52f7045/stable_diffusion.hpp#L385 why？

henbucuoshanghai

I compiled this project with VS 2022 and found the speed of this app is very slower than the app from EdVince.

1

Is it due to the compiler or due to the app code logic changes?

zhanweiw

How could an initial latency space be generated to make img2img?

b4zz4

assets/vocab.txt

2

有什么作用呢？

henbucuoshanghai

missing arugments

1

``` g++ -o test test.cpp -funsafe-math-optimizations -Ofast -flto=auto -funroll-all-loops -pipe -march=native -std=c++20 -Wall -Wextra `pkg-config --cflags --libs ncnn` -lstdc++ -pthread -Wl,--gc-sections -flto -fopt-info-vec-optimized ParserError: Line | 1 | … tra...

xalteropsx

使用GPU加速

1

大佬你好，在win平台怎么用gpu加速呀？

gaoyan2659365465

Int8量化模型

1

发现一个针对SD的[onnx的int8量化脚本](https://github.com/LowinLi/stable-diffusion-streamlit/blob/main/src/stable-diffusion-streamlit/pages/model/quantization.py)和[量化好的库](https://drive.filen.io/f/2d917512-0566-4903-ab55-e3f3415ed18f#MqhCRo5kCbOcaloWuYKlaLTUaDRhUwMd)，这个能转出ncnn的int8模型不？

hubin858130