atomic14 / atomic14/platformio-tensorflow-lite
Error while compiling
- Dominant language
- Dockerfile
- Stars
- 31
- Forks
- 4
- PR merge metrics
- No merged PRs in 30d
Description
Hello!
I've generated tfmicro lib and do all stuff from README but when compiling I got two errors
1. error with including "flatbuffers/flatbuffers.h" , I've fixed it by adding path "[PATH_TO_PROJECT]/lib/tfmicro/third_party/flatbuffers/include". I'm not sure why it's necessary but this line fix the problem. It should be done while copying the library.
2. While compiling I got an error
```
lib/tfmicro/tensorflow/lite/kernels/kernel_util.cc: In function 'std::__cxx11::string tflite::GetShapeDebugString(const TfLiteIntArray*)':
lib/tfmicro/tensorflow/lite/kernels/kernel_util.cc:421:19: error: 'to_string' is not a member of 'std'
str = "[" + std::to_string(shape->data[d]);
^
lib/tfmicro/tensorflow/lite/kernels/kernel_util.cc:423:21: error: 'to_string' is not a member of 'std'
str += ", " + std::to_string(shape->data[d]);
^
*** [.pio/build/esp32cam/libce4/tfmicro/tensorflow/lite/kernels/kernel_util.cc.o] Error 1
```
And this is my c_cpp_properties.json in VSCode project
```
"cStandard": "c99",
"cppStandard": "c++11",
"compilerPath": "/home/kuba/.platformio/packages/toolchain-xtensa32@2.50200.97/bin/xtensa-esp32-elf-gcc",
"compilerArgs": [
"-mfix-esp32-psram-cache-issue",
"-mlongcalls",
""
]
```
I've execute this command `docker build . -t tflite-generator --build-arg tensorflow_version=master` and than follow steps from README.md
Maybe you know how to solve this issue
Regards,
Jakub
Contributor guide
No contributing guide indexed for this repository
Research direction
Reproduce the README.md generation flow with the reported Docker command and inspect the generated lib/tfmicro include paths. Then trace the compilation failures in third_party/flatbuffers/include and tensorflow/lite/kernels/kernel_util.cc using the ESP32 toolchain and the shown c_cpp_properties.json. Done means the generated library builds without either reported error.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp, docker, tensorflow
- Domain
- build-system, embedded-iot
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100