android - 如何构建 TensorFlow Lite 并在 Qt 项目中使用？

Question

我正在尝试启动项目:

 https://mechatronicsblog.com/tensorflow-lite-integration-with-qt-and-v-play-for-multi-platform-machine-learning-apps-on-ios-and-android/

但是有一个问题 - Tensorflow 子模块不正确。

我通过以下命令从主仓库下载了 Tensorflow 库:

git clone https://github.com/tensorflow/tensorflow.git

然后我更新了依赖项:

tensorflow/lite/tools/make/download_dependencies.sh

然后我通过以下命令构建了库:

 bazel build --cxxopt='-D_GLIBCXX_USE_CXX11_ABI=0' -c opt --config=android_arm tensorflow/lite/java:libtensorflowlite_jni

这是 WORKSPACE 的样子:

android_sdk_repository(
    name = "androidsdk",
    api_level = 21,
    build_tools_version = "26.0.2",
    path = "/home/user/Android/SDK",
)

android_ndk_repository(
    name = "androidndk",
    api_level = 20,
    path = "/home/user/android-ndk-r20b",
)

结果是成功的。

然后我复制了库文件夹并将其粘贴到项目文件夹的根目录中。
我没有改变 .pro文件

# TensorFlow Lite - Global
TENSORFLOW_PATH = $$PWD/tensorflow/
TFLITE_MAKE_PATH = $$TENSORFLOW_PATH/tensorflow/lite/tools/make
INCLUDEPATH += $$TENSORFLOW_PATH \
               $$TFLITE_MAKE_PATH/downloads/ \
               $$TFLITE_MAKE_PATH/downloads/eigen \
               $$TFLITE_MAKE_PATH/downloads/gemmlowp \
               $$TFLITE_MAKE_PATH/downloads/neon_2_sse \
               $$TFLITE_MAKE_PATH/downloads/farmhash/src \
               $$TFLITE_MAKE_PATH/downloads/flatbuffers/include

# TensorFlow Lite - Linux
linux:!android {
    INCLUDEPATH += -L$$TFLITE_MAKE_PATH/gen/linux_x86_64/obj
    LIBS += -L$$TFLITE_MAKE_PATH/gen/linux_x86_64/lib/ \
            -ltensorflow-lite -ldl
}

# TensorFlow Lite - Android - armv7a
android {
    QT += androidextras
    LIBS += -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/c \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/core/api \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/kernels \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/kernels/internal \
            -L$$TENSORFLOW_PATH/bazel-bin/external/androidndk \
            -L$$TENSORFLOW_PATH/bazel-bin/external/farmhash_archive \
            -L$$TENSORFLOW_PATH/bazel-bin/external/fft2d \
            -L$$TENSORFLOW_PATH/bazel-bin/external/flatbuffers \
            -lframework -larena_planner -lsimple_memory_arena -lutil -lapi -lc_api_internal \
            -lbuiltin_ops -lbuiltin_op_kernels -lkernel_util -leigen_support -lgemm_support \
            -laudio_utils -lkernel_utils -ltensor_utils -lneon_tensor_utils -lquantization_util \
            -llstm_eval -lstring_util -lcpufeatures -lfarmhash -lfft2d -lflatbuffers
}

# TensorFlow Lite - iOS - Universal library
ios {
        LIBS += -L$$TFLITE_MAKE_PATH/gen/lib/ \
                -framework Accelerate \
                -ltensorflow-lite
}

然后我在以下项目环境中构建它:

Qt Creator 4.11.2
Android NDK 20.1.5948944
Android SDK 26.1.1
Compiler Clang Qt 5.13.2 for Android ARMv7
JVM java-8-oracle

结果失败了。

错误:

/home/user/FelgoProjects/TensorFlowLiteQtVPlay/tensorflow/tensorflow/lite/delegates/nnapi/nnapi_delegate.h:23: error: 'absl/types/optional.h' file not found
#include "absl/types/optional.h"

我想知道如何为 Qt 项目构建和包含 Tensorflow 库。

已编辑:

我编辑了 .pro添加库的文件:

# TensorFlow Lite - Global
TENSORFLOW_PATH = $$PWD/tensorflow/
TFLITE_MAKE_PATH = $$TENSORFLOW_PATH/tensorflow/lite/tools/make
INCLUDEPATH +=  $$TENSORFLOW_PATH \
                $$TFLITE_MAKE_PATH/downloads/ \
                $$TFLITE_MAKE_PATH/downloads/eigen \
                $$TFLITE_MAKE_PATH/downloads/gemmlowp \
                $$TFLITE_MAKE_PATH/downloads/neon_2_sse \
                $$TFLITE_MAKE_PATH/downloads/farmhash/src \
                $$TFLITE_MAKE_PATH/downloads/flatbuffers/include

# TensorFlow Lite - Android - armv7a
android {
    QT += androidextras

    LIBS += -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/c \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/core/api \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/kernels \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/kernels/internal \
            -L$$TENSORFLOW_PATH/bazel-bin/tensorflow/lite/nnapi \
            -L$$TENSORFLOW_PATH/bazel-bin/external/androidndk \
            -L$$TENSORFLOW_PATH/bazel-bin/external/farmhash_archive \
            -L$$TENSORFLOW_PATH/bazel-bin/external/fft2d \
            -L$$TENSORFLOW_PATH/bazel-bin/external/flatbuffers \
            -L$$TENSORFLOW_PATH/bazel-bin/external/flatbuffers/src \
            -L$$TENSORFLOW_PATH/bazel-bin/external/ruy/ruy \
            -L$$TENSORFLOW_PATH/bazel-bin/external/ruy/ruy/profiler \
            -lallocation.pic -larena_planner.pic -larena_planner.pic -lminimal_logging.pic \
            -lsimple_memory_arena.pic -lstring_util.pic -lutil.pic \
            -lapi.pic -lbuiltin_op_kernels.pic -lbuiltin_ops.pic -lcpu_backend_context.pic -lcpu_backend_gemm.pic -leigen_support.pic \
            -lkernel_util.pic -llstm_eval.pic -laudio_utils.pic -lkernel_utils.pic -lneon_tensor_utils.pic \
            -lportable_tensor_utils.pic -ltensor_utils.pic -lquantization_util.pic -ltranspose_utils.pic \
            -lfarmhash.pic -lfft2d.pic -lflatbuffers.pic \
            -lallocator.pic -lapply_multiplier.pic -lblocking_counter.pic -lblock_map.pic -lcontext.pic -lcontext_get_ctx.pic \
            -lctx.pic -ldetect_arm.pic -ldetect_x86.pic -lhave_built_path_for_avx2.pic -lhave_built_path_for_avx512.pic \
            -lhave_built_path_for_avxvnni.pic -lhave_built_path_for_sse42.pic -lkernel_arm.pic -lkernel_avx2.pic \
            -lkernel_avx512.pic -lkernel_avxvnni.pic -lkernel_sse42.pic -lpack_arm.pic -lpack_avx2.pic -lpack_avx512.pic \
            -lpack_avxvnni.pic -lpack_sse42.pic -lprepacked_cache.pic -lthread_pool.pic -ltrace.pic -ltrmul.pic \
            -ltune.pic -lwait.pic -linstrumentation.pic -lnnapi_implementation.pic -lnnapi_util.pic
}

编译后有一些错误。

tensorflow/lite/util.cc:47: error: undefined reference to 'TfLiteIntArrayCreate'
tensorflow/lite/util.cc:47: error: undefined reference to 'TfLiteIntArrayCreate'
tensorflow/lite/kernels/activations.cc:265: error: undefined reference to 'TfLiteIntArrayCopy'
tensorflow/lite/kernels/activations.cc:291: error: undefined reference to 'TfLiteIntArrayCopy'
tensorflow/lite/kernels/activations.cc:380: error: undefined reference to 'TfLiteIntArrayCopy'
tensorflow/lite/kernels/activations.cc:612: error: undefined reference to 'TfLiteIntArrayCopy'
tensorflow/lite/kernels/activations.cc:729: error: undefined reference to 'TfLiteTypeGetName'
tensorflow/lite/kernels/activations.cc:757: error: undefined reference to 'TfLiteTypeGetName'
tensorflow/lite/kernels/activations.cc:840: error: undefined reference to 'TfLiteTypeGetName'
tensorflow/lite/kernels/activations.cc:1084: error: undefined reference to 'TfLiteTypeGetName'
tensorflow/lite/kernels/arg_min_max.cc:40: error: undefined reference to 'TfLiteIntArrayCreate'
tensorflow/lite/kernels/arg_min_max.cc:40: error: undefined reference to 'TfLiteIntArrayCreate'
tensorflow/lite/kernels/basic_rnn.cc:104: error: undefined reference to 'TfLiteIntArrayFree'
tensorflow/lite/kernels/basic_rnn.cc:110: error: undefined reference to 'TfLiteIntArrayEqual'
tensorflow/lite/kernels/basic_rnn.cc:120: error: undefined reference to 'TfLiteIntArrayEqual'
tensorflow/lite/kernels/basic_rnn.cc:133: error: undefined reference to 'TfLiteIntArrayEqualsArray'
tensorflow/lite/kernels/basic_rnn.cc:144: error: undefined reference to 'TfLiteIntArrayEqualsArray'
tensorflow/lite/kernels/basic_rnn.cc:157: error: undefined reference to 'TfLiteIntArrayEqualsArray'
tensorflow/lite/kernels/basic_rnn.cc:168: error: undefined reference to 'TfLiteIntArrayEqualsArray'
tensorflow/lite/kernels/batch_matmul.cc:108: error: undefined reference to 'TfLiteIntArrayFree'
tensorflow/lite/kernels/bidirectional_sequence_lstm.cc:527: error: undefined reference to 'TfLiteIntArrayFree'
tensorflow/lite/kernels/bidirectional_sequence_lstm.cc:527: error: undefined reference to 'TfLiteIntArrayFree'
tensorflow/lite/kernels/bidirectional_sequence_lstm.cc:630: error: undefined reference to 'TfLiteIntArrayEqual'
tensorflow/lite/kernels/bidirectional_sequence_lstm.cc:642: error: undefined reference to 'TfLiteIntArrayEqual'
./tensorflow/lite/kernels/internal/reference/densify.h:36: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<float>::FormatConverter(std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, TfLiteSparsity const&)'
./tensorflow/lite/kernels/internal/reference/densify.h:38: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<float>::SparseToDense(float const*)'
./tensorflow/lite/kernels/internal/reference/densify.h:36: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<signed char>::FormatConverter(std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, TfLiteSparsity const&)'
./tensorflow/lite/kernels/internal/reference/densify.h:38: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<signed char>::SparseToDense(signed char const*)'
tensorflow/lite/kernels/embedding_lookup_sparse.cc:178: error: undefined reference to 'TfLiteTensorRealloc'
tensorflow/lite/kernels/expand_dims.cc:105: error: undefined reference to 'TfLiteTensorRealloc'
./tensorflow/lite/kernels/internal/reference/sparse_ops/fully_connected.h:35: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<float>::FormatConverter(std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, TfLiteSparsity const&)'
./tensorflow/lite/kernels/internal/reference/sparse_ops/fully_connected.h:37: error: undefined reference to 'tflite::optimize::sparsity::FormatConverter<float>::SparseToDense(float const*)'
tensorflow/lite/kernels/if.cc:85: error: undefined reference to 'tflite::impl::Subgraph::ResizeInputTensor(int, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&)'
tensorflow/lite/kernels/if.cc:92: error: undefined reference to 'tflite::impl::Subgraph::AllocateTensors()'
tensorflow/lite/kernels/if.cc:85: error: undefined reference to 'tflite::impl::Subgraph::ResizeInputTensor(int, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&)'
tensorflow/lite/kernels/if.cc:92: error: undefined reference to 'tflite::impl::Subgraph::AllocateTensors()'
tensorflow/lite/kernels/if.cc:155: error: undefined reference to 'tflite::impl::Subgraph::Invoke()'
tensorflow/lite/kernels/reshape.cc:156: error: undefined reference to 'TfLiteTensorRealloc'
tensorflow/lite/kernels/while.cc:152: error: undefined reference to 'tflite::impl::Subgraph::AllocateTensors()'
tensorflow/lite/kernels/while.cc:170: error: undefined reference to 'tflite::impl::Subgraph::AllocateTensors()'
tensorflow/lite/kernels/while.cc:57: error: undefined reference to 'tflite::impl::Subgraph::ResizeInputTensor(int, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&)'
tensorflow/lite/kernels/while.cc:267: error: undefined reference to 'tflite::impl::Subgraph::Invoke()'
tensorflow/lite/kernels/while.cc:292: error: undefined reference to 'tflite::impl::Subgraph::Invoke()'
tensorflow/lite/kernels/while.cc:57: error: undefined reference to 'tflite::impl::Subgraph::ResizeInputTensor(int, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&)'
tensorflow/lite/kernels/register.cc:34: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:35: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:36: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int, int)'
tensorflow/lite/kernels/register.cc:38: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:39: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int, int)'
tensorflow/lite/kernels/register.cc:41: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int, int)'
tensorflow/lite/kernels/register.cc:43: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int, int)'
tensorflow/lite/kernels/register.cc:52: error: undefined reference to 'tflite::MutableOpResolver::AddBuiltin(tflite::BuiltinOperator, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:291: error: undefined reference to 'tflite::MutableOpResolver::AddCustom(char const*, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:294: error: undefined reference to 'tflite::MutableOpResolver::AddCustom(char const*, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:295: error: undefined reference to 'tflite::MutableOpResolver::AddCustom(char const*, TfLiteRegistration const*, int)'
tensorflow/lite/kernels/register.cc:297: error: undefined reference to 'tflite::MutableOpResolver::AddCustom(char const*, TfLiteRegistration const*, int)'
external/androidndk/ndk/sources/cxx-stl/llvm-libc++/include/unordered_map:0: error: undefined reference to 'vtable for tflite::MutableOpResolver'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
./tensorflow/lite/kernels/register.h:0: error: undefined reference to 'vtable for tflite::MutableOpResolver'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
/home/sergey/FelgoProjects/TensorFlowLiteQtVPlay/tensorflow//bazel-bin/tensorflow/lite/kernels/libbuiltin_ops.pic.a(register.pic.o):register.cc:vtable for tflite::ops::builtin::BuiltinOpResolver: error: undefined reference to 'tflite::MutableOpResolver::FindOp(tflite::BuiltinOperator, int) const'
/home/sergey/FelgoProjects/TensorFlowLiteQtVPlay/tensorflow//bazel-bin/tensorflow/lite/kernels/libbuiltin_ops.pic.a(register.pic.o):register.cc:vtable for tflite::ops::builtin::BuiltinOpResolver: error: undefined reference to 'tflite::MutableOpResolver::FindOp(char const*, int) const'
/home/sergey/FelgoProjects/TensorFlowLiteQtVPlay/tensorflow//bazel-bin/tensorflow/lite/kernels/libbuiltin_ops.pic.a(register.pic.o):register.cc:typeinfo for tflite::ops::builtin::BuiltinOpResolver: error: undefined reference to 'typeinfo for tflite::MutableOpResolver'
../../Felgo/Felgo/android_armv7/include/QtQml/qqml.h:0: error: undefined reference to 'ObjectsRecogFilter::staticMetaObject'
../../Felgo/Felgo/android_armv7/include/QtQml/qqml.h:0: error: undefined reference to 'ObjectsRecogFilter::staticMetaObject'
../TensorFlowLiteQtVPlay/auxutils.h:0: error: undefined reference to 'vtable for AuxUtils'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../../Felgo/Felgo/android_armv7/include/QtCore/qmetatype.h:0: error: undefined reference to 'ObjectsRecogFilter::staticMetaObject'
../../Felgo/Felgo/android_armv7/include/QtCore/qmetatype.h:0: error: undefined reference to 'ObjectsRecogFilter::staticMetaObject'
../TensorFlowLiteQtVPlay/objectsrecogfilter.h:0: error: undefined reference to 'vtable for ObjectsRecogFilter'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflowthread.h:0: error: undefined reference to 'vtable for TensorflowThread'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflowthread.h:0: error: undefined reference to 'vtable for WorkerTF'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflow/tensorflow/lite/mutable_op_resolver.h:0: error: undefined reference to 'vtable for tflite::MutableOpResolver'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
/home/sergey/android-ndk-r20b/sources/cxx-stl/llvm-libc++/include/memory:2339: error: undefined reference to 'tflite::FlatBufferModel::~FlatBufferModel()'
/home/sergey/android-ndk-r20b/sources/cxx-stl/llvm-libc++/include/memory:2339: error: undefined reference to 'tflite::impl::Interpreter::~Interpreter()'
main.o:main.cpp:vtable for QQmlPrivate::QQmlElement<ObjectsRecogFilter>: error: undefined reference to 'ObjectsRecogFilter::metaObject() const'
main.o:main.cpp:vtable for QQmlPrivate::QQmlElement<ObjectsRecogFilter>: error: undefined reference to 'ObjectsRecogFilter::qt_metacast(char const*)'
main.o:main.cpp:vtable for QQmlPrivate::QQmlElement<ObjectsRecogFilter>: error: undefined reference to 'ObjectsRecogFilter::qt_metacall(QMetaObject::Call, int, void**)'
main.o:main.cpp:typeinfo for QQmlPrivate::QQmlElement<ObjectsRecogFilter>: error: undefined reference to 'typeinfo for ObjectsRecogFilter'
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:16: error: undefined reference to 'ObjectsRecogFilter::initializedChanged(bool const&)'
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:0: error: undefined reference to 'vtable for ObjectsRecogFilter'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:161: error: undefined reference to 'ObjectsRecogFilter::runTensorFlow(QImage)'
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:300: error: undefined reference to 'ObjectsRecogFilter::initializedChanged(bool const&)'
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:248: error: undefined reference to 'ObjectsRecogFilter::initializedChanged(bool const&)'
../TensorFlowLiteQtVPlay/objectsrecogfilter.cpp:276: error: undefined reference to 'ObjectsRecogFilter::initializedChanged(bool const&)'
../TensorFlowLiteQtVPlay/tensorflowthread.cpp:16: error: undefined reference to 'WorkerTF::results(int, QStringList, QList<double>, QList<QRectF>, double)'
../TensorFlowLiteQtVPlay/tensorflowthread.cpp:17: error: undefined reference to 'WorkerTF::finished()'
../TensorFlowLiteQtVPlay/tensorflowthread.cpp:0: error: undefined reference to 'vtable for TensorflowThread'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflowthread.cpp:47: error: undefined reference to 'TensorflowThread::results(int, QStringList, QList<double>, QList<QRectF>, double)'
../TensorFlowLiteQtVPlay/tensorflowthread.h:0: error: undefined reference to 'vtable for WorkerTF'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:109: error: undefined reference to 'tflite::FlatBufferModel::BuildFromFile(char const*, tflite::ErrorReporter*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:118: error: undefined reference to 'tflite::impl::InterpreterBuilder::InterpreterBuilder(tflite::FlatBufferModel const&, tflite::OpResolver const&)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:121: error: undefined reference to 'tflite::impl::InterpreterBuilder::operator()(std::__ndk1::unique_ptr<tflite::impl::Interpreter, std::__ndk1::default_delete<tflite::impl::Interpreter> >*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:128: error: undefined reference to 'tflite::impl::Interpreter::UseNNAPI(bool)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:131: error: undefined reference to 'tflite::impl::Interpreter::SetNumThreads(int)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:136: error: undefined reference to 'tflite::impl::Interpreter::AllocateTensors()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:166: error: undefined reference to 'tflite::impl::InterpreterBuilder::~InterpreterBuilder()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:166: error: undefined reference to 'tflite::impl::InterpreterBuilder::~InterpreterBuilder()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:272: error: undefined reference to 'tflite::impl::Interpreter::Invoke()'
../TensorFlowLiteQtVPlay/tensorflow/tensorflow/lite/stderr_reporter.h:0: error: undefined reference to 'vtable for tflite::StderrReporter'
/home/sergey/android-ndk-r20b/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/lib/gcc/arm-linux-androideabi/4.9.x/../../../../arm-linux-androideabi/bin/ld: the vtable symbol may be undefined because the class is missing its key function
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:316: error: undefined reference to 'tflite::DefaultErrorReporter()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:316: error: undefined reference to 'tflite::impl::Interpreter::Interpreter(tflite::ErrorReporter*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:321: error: undefined reference to 'tflite::impl::Interpreter::AddTensors(int, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:324: error: undefined reference to 'tflite::impl::Interpreter::AddTensors(int, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:327: error: undefined reference to 'tflite::impl::Interpreter::SetInputs(std::__ndk1::vector<int, std::__ndk1::allocator<int> >)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:328: error: undefined reference to 'tflite::impl::Interpreter::SetOutputs(std::__ndk1::vector<int, std::__ndk1::allocator<int> >)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:337: error: undefined reference to 'tflite::MutableOpResolver::FindOp(tflite::BuiltinOperator, int) const'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:340: error: undefined reference to 'tflite::impl::Interpreter::AddNodeWithParameters(std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, char const*, unsigned int, void*, TfLiteRegistration const*, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:341: error: undefined reference to 'tflite::impl::Interpreter::AllocateTensors()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:354: error: undefined reference to 'tflite::impl::Interpreter::Invoke()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:316: error: undefined reference to 'tflite::DefaultErrorReporter()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:316: error: undefined reference to 'tflite::impl::Interpreter::Interpreter(tflite::ErrorReporter*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:321: error: undefined reference to 'tflite::impl::Interpreter::AddTensors(int, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:324: error: undefined reference to 'tflite::impl::Interpreter::AddTensors(int, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:327: error: undefined reference to 'tflite::impl::Interpreter::SetInputs(std::__ndk1::vector<int, std::__ndk1::allocator<int> >)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:328: error: undefined reference to 'tflite::impl::Interpreter::SetOutputs(std::__ndk1::vector<int, std::__ndk1::allocator<int> >)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:337: error: undefined reference to 'tflite::MutableOpResolver::FindOp(tflite::BuiltinOperator, int) const'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:340: error: undefined reference to 'tflite::impl::Interpreter::AddNodeWithParameters(std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, std::__ndk1::vector<int, std::__ndk1::allocator<int> > const&, char const*, unsigned int, void*, TfLiteRegistration const*, int*)'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:341: error: undefined reference to 'tflite::impl::Interpreter::AllocateTensors()'
../TensorFlowLiteQtVPlay/tensorflowlite.cpp:354: error: undefined reference to 'tflite::impl::Interpreter::Invoke()'
../TensorFlowLiteQtVPlay/tensorflow/tensorflow/lite/interpreter.h:178: error: undefined reference to 'tflite::impl::Interpreter::SetTensorParametersReadWrite(int, TfLiteType, char const*, unsigned int, int const*, TfLiteQuantizationParams, bool, unsigned int, int const*)'
tensorflow/lite/string_util.cc:108: error: undefined reference to 'TfLiteTensorReset'
clang++: error: linker command failed with exit code 1 (use -v to see invocation)
make: *** [libTensorFlowLiteQtVPlay.so] Error 1

Answer 1

我是本教程的作者。你能尝试用 编译TensorFlow吗？ libstdc++ 通过使用此命令 as described in the tutorial :

bazel build --cxxopt='--std=c++11' -c opt --config=android_arm tensorflow/lite/java:libtensorflowlite_jni

也许您的问题与 mixing libc++ and libstdc++ 有关, Qt 使用 libstdc++ 你正在用 编译 TensorFlow libc++ 在你的 bazel 命令中。

让我知道这是否有帮助。