Installing Tensorflow Lite with CMake

user_16197261 0
Viewed 237

I am trying to install Tensorflow Lite on Raspberry Pi Zero W. I followed the steps from the official website (https://www.tensorflow.org/lite/guide/build_cmake_arm). But I get the following error message:

-- The C compiler identification is unknown
-- The CXX compiler identification is unknown
-- Check for working C compiler: /home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc

-- Check for working C compiler: /home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc -- broken

CMake Error at /usr/share/cmake-3.16/Modules/CMakeTestCCompiler.cmake:60 (message):

The C compiler

"/home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc"

is not able to compile a simple test program.

It fails with the following output:

Change Dir: /home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeTmp

Run Build Command(s):/usr/bin/make cmTC_ed1a3/fast && /usr/bin/make -f CMakeFiles/cmTC_ed1a3.dir/build.make CMakeFiles/cmTC_ed1a3.dir/build
make[1]: Entering directory '/home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeTmp'
Building C object CMakeFiles/cmTC_ed1a3.dir/testCCompiler.c.o
/home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc   -march=armv6 -mfpu=vfp -funsafe-math-optimizations    -o CMakeFiles/cmTC_ed1a3.dir/testCCompiler.c.o   -c /home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeTmp/testCCompiler.c
/home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc: 1: /home/pi/toolchains/arm-rpi-linux-gnueabihf/x64-gcc-6.5.0/arm-rpi-linux-gnueabihf/bin/arm-rpi-linux-gnueabihf-gcc: Syntax error: ")" unexpected
make[1]: *** [CMakeFiles/cmTC_ed1a3.dir/build.make:66: CMakeFiles/cmTC_ed1a3.dir/testCCompiler.c.o] Error 2
make[1]: Leaving directory '/home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeTmp'
make: *** [Makefile:121: cmTC_ed1a3/fast] Error 2

CMake will not be able to correctly generate this project.

Call Stack (most recent call first):

CMakeLists.txt:44 (project)

Configuring incomplete, errors occurred!
See also "/home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeOutput.log".
See also "/home/pi/Project/cmake-3.16.0/CMakeFiles/CMakeError.log".

Does anyone know how to fix the problem?

raspberry-pi-zero tensorflow cmake linux c
1 Answers

I ran into the same problem like you! The instructions given at https://www.tensorflow.org/lite/guide/build_cmake_arm?hl=en are not intended to be executed on the Raspberry Pi (at least the last ones). Let me share, what I'll figured out.

Short answer: you're trying to run a cross-compile of TensorFlow Lite for an ARM made for x86 platform on an ARM system (your Raspberry Pi). What you're trying to do actually needs to be done on an x86 platform. (on the top of the tutorial page is a hint: "following instructions have been tested on Ubuntu 16.04.3 64-bit PC (AMD64)" which reveals that the commands have been run on x86)

Some details: The message ‘Syntax error: "(" unexpected’ probably comes from a shell. Why is a shell trying to execute your binary? Because the kernel tried and gave up. Why did the kernel fail to run the program? Because the executable is not in a supported format -> meant to be run on x86 but was executed on ARM. https://unix.stackexchange.com/questions/336436/syntax-error-unexpected-when-execute-a-compiled-c-program

With older TensorFlow versions it was possible to compile TF lite on the Raspberry Pi itself via the ./tensorflow-2.4.1/tensorflow/lite/tools/make/download_dependencies.sh /and build_aarch64_lib.sh scripts. However, this scripts got removed in version 2.8 (or even with an earlier version before). See note in the Readme of the Git: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/tools/make/README.md

Using Python for inference
Unfortunately I have no real solution how to compile / built TensorFlow Lite on the Raspberry itself. I personally ended up using Python (only needed inference) and followed this Tutorial: https://qengineering.eu/install-tensorflow-2.7-on-raspberry-64-os.html

user_133206770
Related