compiling in raspi 4

[RorschachTest]

trying to build on raspi 4 with rasbain os 64 bit everything works fine until in try. "cmake --build . --config Release" command for OnnxStream Not sure what i am doing wrong,Please help

Os version PRETTY_NAME="Debian GNU/Linux 11 (bullseye)"

LOGS: (diffusionpi) rihaz@pi:~/diffusionpi/OnnxStream/src/build $ cmake --build . --config Release [ 33%] Building CXX object CMakeFiles/sd.dir/sd.cpp.o [ 66%] Building CXX object CMakeFiles/sd.dir/onnxstream.cpp.o In file included from /usr/include/c++/10/bits/stl_algobase.h:64, from /usr/include/c++/10/vector:60, from /home/rihaz/diffusionpi/OnnxStream/src/onnxstream.h:4, from /home/rihaz/diffusionpi/OnnxStream/src/onnxstream.cpp:1: /usr/include/c++/10/bits/stl_pair.h: In instantiation of 'constexpr std::pair<typename std::__strip_reference_wrapper<typename std::decay<_Tp>::type>::__type, typename std::__strip_reference_wrapper<typename std::decay<_Tp2>::type>::__type> std::make_pair(_T1&&, _T2&&) [with _T1 = float&; _T2 = float&; typename std::__strip_reference_wrapper<typename std::decay<_Tp2>::type>::__type = float; typename std::__strip_reference_wrapper<typename std::decay<_Tp>::type>::__type = float]': /home/rihaz/diffusionpi/OnnxStream/src/onnxstream.cpp:2085:60: required from here /usr/include/c++/10/bits/stl_pair.h:567:5: note: parameter passing for argument of type 'std::pair<float, float>' when C++17 is enabled changed to match C++14 in GCC 10.1 567 | make_pair(_T1&& __x, _T2&& __y) | ^~~~~~~~~ [100%] Linking CXX executable sd [100%] Built target sd

(diffusionpi) rihaz@pi:~/diffusionpi/OnnxStream/src/build $ ./sd --rpi --steps 3 --output result.png ----------------[start]------------------ positive_prompt: a photo of an astronaut riding a horse on mars negative_prompt: ugly, blurry output_png_path: result.png steps: 3 ----------------[prompt]------------------ === ERROR === unable to open file: ./tokenizer/vocab.txt

[vitoplantamura]

hi,

you should download the Stable Diffusion weights before running the software. The path of the weights can be specified via the "--models-path" option when running the software. The weights can be downloaded here:

https://github.com/vitoplantamura/OnnxStream/releases/tag/v0.1

Thanks, Vito

[atomicrajat]

Hi, First of all its a fascinating project, after encountering this repo i immediately took my rpi and wanted to try it. I was able to run the project on raspberry pi 4 8GB board running Raspberry Pi OS Lite (64 bit) So just wanted to update on the performance so that others can get an idea if they are running the same.

Steps: 7 (results in better generated output) Time for each step: 192938 ms which is 3 min 21 sec

Thanks, Rajat

[dstab001]

I want to try this on my pi4 8gb. What needs to be changed in the build? Just learning

[vitoplantamura]

@dstab001, just follow the instructions in the main README :-)