Tensorflowlite移植ARM平台iMX6

一、LINUX环境下操作：

    1.安装交叉编译SDK （仅针对该型号：i.MX6,不同芯片需要对应的交叉编译SDK）

            编译方法参考：手动编译用于i.MX6系列的交叉编译SDK

     2.下载Tensorflow

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

       cd tensorflow

       git checkout r1.10

      Tensorflow与Bazel编译器（及CUDA，CUDNN）之间需要对应，否则会有兼容性问题。

          tensorflowr1.10     python 2.7,3.6   Bazel:0.18.0-0.19.2

          tensorflowr1.12     python 2.7,3.6   Bazel:0.18.0-0.19.2

          tensorflowr1.14     python 2.7,3.6   Bazel:0.24.0 - 0.25.2

      3、下载并安装编译工具Bazel

          安装依赖包：

                sudo apt-get install pkg-config zip g++ zilb1g-dev unzip

          下载Bazel包：

                wget https://github.com/bazelbuild/bazel/releases/download/0.18.1/bazel-0.181-installer-linux-86_64.sh

          安装Bazel：

                 chmod +x bazel-0.18.1-installer-linux-86_64.sh

                 ./bazel-0.18.1-installer-linux-86_64.sh --user

          设置环境变量：

                 sudo vi ~/.bashrc，在文件最后添加：export PATH=$PATH":~/bin"

                 source ~/.bashrc

    （如果仅仅是测试DEMO在ARM板上使用，可直接跳过4，5，6，7，8步，直接进行第9步）

    4、编译配置：

          在Tensorflow源码根目录运行：

             ./configure     （编译LINUX平台时使用默认设置：-march=native，编译ARM平台时需设置成相应值：-march=armv7-a）

     5、编译pip:

          bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package

     7、编译包：

          ./bazel-bin/tensorflow/tools/pip__package/build_pip_package /tmp/tensorflow_pkg

     8、安装包：

          pip install /tmp/tensorflow_pkg/tensorflow-version-tags.whl

     9、下载依赖库：

         ./tensorflow/contrib/lite/tools/make/download_dependencies.sh(不同版本，位置略有不同，本文路径为r1.10版本)

     10、编译Tensorflow Lite:

         方法1：（参考网上的，但我调试了几天，并未生成该库，可行性待验证！）

                目前Tensorflow仅支持树莓派： ./tensorflow/contrib/lite/tools/make/build_rpi_lib.sh，该脚本的目标编译平台是ARMv7，即使目标平台是ARMv8，也不要更改。因为设置编译平台为ARMv7可以优化编译，提高运行速度。

                所以，需要针对iMX6，制作一份build_imx6_lib.sh

set -e 

SCRIPT_DIR="$(cd "$(dirname "$(BASH_SOURCE[0]}")" && pwd)"
cd "$SCRIPT_DIR/../../.."

CC_PREFIX=arm-poky-linux-gnueabi- make -j 3 -f tensorflow/contrib/lite/Makefile TARGET-imx6 TARGET_ARCH=armv7-a

                生成静态库位置为：

               ./tensorflow/contrib/lite/toos/make/gen/rpi_arm7l/lib/libtensorflow-lite.a静态库。

         方法2：（使用ARM（iMX6）交叉编译包，编译出来的包可以用于iMX6平台）

                1.打开Tensorflow/contrib/lite/kernels/internal/BUILD

config_setting(
       name = "armv7a",
       values = {"cpu":"armv7a",},
)

"armv7a":[
    "-O3",
    "-mfpu=vfpv3",
    "-mfloat-abi=hard",
],

add code above to NEON_FLAGS_IF_APPLICABLE


":armv6":[

    ":neon_tensor_utils",
    ],
":armv7a":[
    ":neon_tensor_utils",
],

add code above to cc_library  tensor_utils select options

                2.在根目录下创建一个文件：build_armv7_tflite.sh

#!/bin/bash

bazel build --copt="-fPIC" --copt="-march=armv6" 
     --copt="-mfloat-abi=hard" 
     --copt="-mfpu=vfpv3" 
    --copt="-funsafe-math-optimizations" 
     --copt="-Wno-unused-function" 
     --copt="-Wno-sign-compare" 
     --copt="-ftree-vectorize" 
     --copt="-fomit-frame-pointer" 
     --cxxopt='--std=c++11' 
     --cxxopt="-Wno-maybe-uninitialized" 
     --cxxopt="-Wno-narrowing" 
     --cxxopt="-Wno-unused" 
     --cxxopt="-Wno-comment" 
     --cxxopt="-Wno-unused-function" 
     --cxxopt="-Wno-sign-compare" 
     --cxxopt="-funsafe-math-optimizations" 
     --linkopt="-lstdc++" 
     --linkopt="-mfloat-abi=hard" 
     --linkopt="-mfpu=vfpv3" 
     --linkopt="-funsafe-math-optimizations" 
     --verbose_failures 
     --strip=always 
     --crosstool_top=//armv6-compiler:toolchain --cpu=armv7 --config=opt 
33     tensorflow/contrib/lite/examples/label_image

                3.编译该文件build_armv7_tflite.sh，会碰到一个错误：.../.../read-ld:unrecognized options : --icf=all

                   解决方法：找到文件./build_def.bzl ,打开，去除所有--icf=all标识的信息即可。

                   编译后，会在bazel_bin下生成指定的文件label_image。

                4.生成头和库文件：      

                    libbuiltin_ops.a libframework.a libneon_tensor_utils.a libquantization_util.a libtensor_utils.a

                    libcontext.a libfarmhash.a libgemm_support.a libportable_tensor_utils.a  libstring_util.a

     11、编译模型：

          默认情况下label_image并未编译进去，需要修改Makefile，可参考minimal APK,主要修改以下三部分内容：

            LABELIMAGE_SRCS

            LABELIMAGE_BINARY

            LABEL_OBJS

# Find where we're running from, so we can store generated files here.
ifeq ($(origin MAKEFILE_DIR), undefined)
    MAKEFILE_DIR := $(shell dirname $(realpath $(lastword $(MAKEFILE_LIST))))
endif

# Try to figure out the host system
HOST_OS :=
ifeq ($(OS),Windows_NT)
    HOST_OS = WINDOWS
else
    UNAME_S := $(shell uname -s)
    ifeq ($(UNAME_S),Linux)
            HOST_OS := LINUX
    endif
    ifeq ($(UNAME_S),Darwin)
        HOST_OS := OSX
    endif
endif

#HOST_ARCH := $(shell if [[ $(shell uname -m) =~ i[345678]86 ]]; then echo x86_32; else echo $(shell uname -m); fi)

# Self-hosting
TARGET_ARCH := ${HOST_ARCH}
CROSS := imx6
$(warning "CROSS :$(CROSS) HOST_ARCH:$(HOST_ARCH) TARGET_ARCH:$(TARGET_ARCH) TARGET_TOOLCHAIN_PREFIX:$(TARGET_TOOLCHAIN_PREFIX)")
# Cross compiling
ifeq ($(CROSS),imx6)
  TARGET_ARCH := armv7-a
  TARGET_TOOLCHAIN_PREFIX := arm-poky-linux-gnueabi-
endif

ifeq ($(CROSS),rpi)
  TARGET_ARCH := armv7l
  TARGET_TOOLCHAIN_PREFIX := arm-linux-gnueabihf-
endif

ifeq ($(CROSS),riscv)
  TARGET_ARCH := riscv
  TARGET_TOOLCHAIN_PREFIX := riscv32-unknown-elf-
endif
ifeq ($(CROSS),stm32f7)
  TARGET_ARCH := armf7
  TARGET_TOOLCHAIN_PREFIX := arm-none-eabi-
endif
ifeq ($(CROSS),stm32f1)
  TARGET_ARCH := armm1
  TARGET_TOOLCHAIN_PREFIX := arm-none-eabi-
endif

# Where compiled objects are stored.
OBJDIR := $(MAKEFILE_DIR)/gen/obj/
BINDIR := $(MAKEFILE_DIR)/gen/bin/
LIBDIR := $(MAKEFILE_DIR)/gen/lib/
GENDIR := $(MAKEFILE_DIR)/gen/obj/

LIBS :=
ifeq ($(TARGET_ARCH),x86_64)
        CXXFLAGS += -fPIC -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -pthread # -msse4.2
endif


ifeq ($(TARGET_ARCH),armv7-a)
        CXXFLAGS += -pthread -fPIC
    LIBS += -ldl
endif

ifeq ($(TARGET_ARCH),armv7l)
        CXXFLAGS += -mfpu=neon -pthread -fPIC
    LIBS += -ldl
endif

ifeq ($(TARGET_ARCH),riscv)
#        CXXFLAGS += -march=gap8
        CXXFLAGS += -DTFLITE_MCU
    LIBS += -ldl
    BUILD_TYPE := micro
endif

ifeq ($(TARGET_ARCH),armf7)
        CXXFLAGS += -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -DTFLITE_MCU
        CXXFLAGS += -fno-rtti -fmessage-length=0 -fno-exceptions -fno-builtin -ffunction-sections -fdata-sections
        CXXFLAGS += -funsigned-char -MMD
        CXXFLAGS += -mcpu=cortex-m7 -mthumb -mfpu=fpv5-sp-d16 -mfloat-abi=softfp
        CXXFLAGS += '-std=gnu++11' '-fno-rtti' '-Wvla' '-c' '-Wall' '-Wextra' '-Wno-unused-parameter' '-Wno-missing-field-initializers' '-fmessage-length=0' '-fno-exceptions' '-fno-builtin' '-ffunction-sections' '-fdata-sections' '-funsigned-char' '-MMD' '-fno-delete-null-pointer-checks' '-fomit-frame-pointer' '-Os'
    LIBS += -ldl
    BUILD_TYPE := micro
endif
ifeq ($(TARGET_ARCH),armm1)
        CXXFLAGS += -DGEMMLOWP_ALLOW_SLOW_SCALAR_FALLBACK -mcpu=cortex-m1 -mthumb -DTFLITE_MCU
        CXXFLAGS += -fno-rtti -fmessage-length=0 -fno-exceptions -fno-builtin -ffunction-sections -fdata-sections
        CXXFLAGS += -funsigned-char -MMD
    LIBS += -ldl
endif

# Settings for the host compiler.
#CXX := $(CC_PREFIX) ${TARGET_TOOLCHAIN_PREFIX}g++
#CXX := ${TARGET_TOOLCHAIN_PREFIX}g++ -march=armv7-a -mfpu=neon  -mfloat-abi=hard -mcpu=cortex-a9 --sysroot=/opt/fsl-imx-x11/4.1.15-2.1.0/sysroots/x86_64-pokysdk-linux/usr/bin/arm-poky-linux-gnueabi 
CXXFLAGS += -O3 -DNDEBUG
CCFLAGS := ${CXXFLAGS}
#CXXFLAGS += --std=c++11
CXXFLAGS += --std=c++0x     #wly
#CC := ${TARGET_TOOLCHAIN_PREFIX}gcc
#AR := ${TARGET_TOOLCHAIN_PREFIX}ar
CFLAGS :=
LDOPTS :=
LDOPTS += -L/usr/local/lib
ARFLAGS := -r

INCLUDES := 
-I. 
-I$(MAKEFILE_DIR)/../../../ 
-I$(MAKEFILE_DIR)/downloads/ 
-I$(MAKEFILE_DIR)/downloads/eigen 
-I$(MAKEFILE_DIR)/downloads/gemmlowp 
-I$(MAKEFILE_DIR)/downloads/neon_2_sse 
-I$(MAKEFILE_DIR)/downloads/farmhash/src 
-I$(MAKEFILE_DIR)/downloads/flatbuffers/include 
-I$(GENDIR)
# This is at the end so any globally-installed frameworks like protobuf don't
# override local versions in the source tree.
#INCLUDES += -I/usr/local/include
INCLUDES += -I/usr/include
LIBS += 
-lstdc++ 
-lpthread 
-lm 
-lz

# If we're on Linux, also link in the dl library.
ifeq ($(HOST_OS),LINUX)
    LIBS += -ldl
endif

include $(MAKEFILE_DIR)/ios_makefile.inc
ifeq ($(CROSS),rpi)
#include $(MAKEFILE_DIR)/rpi_makefile.inc
endif
ifeq ($(CROSS),imx6)
include $(MAKEFILE_DIR)/imx6_makefile.inc
endif
# This library is the main target for this makefile. It will contain a minimal
# runtime that can be linked in to other programs.
LIB_NAME := libtensorflow-lite.a
LIB_PATH := $(LIBDIR)$(LIB_NAME)

# A small example program that shows how to link against the library.
MINIMAL_PATH := $(BINDIR)minimal
LABEL_IMAGE_PATH :=$(BINDIR)label_image

MINIMAL_SRCS := 
tensorflow/contrib/lite/examples/minimal/minimal.cc
MINIMAL_OBJS := $(addprefix $(OBJDIR), 
$(patsubst %.cc,%.o,$(patsubst %.c,%.o,$(MINIMAL_SRCS))))


LABEL_IMAGE_SRCS := 
tensorflow/contrib/lite/examples/label_image/label_image.cc 
tensorflow/contrib/lite/examples/label_image/bitmap_helpers.cc 
LABEL_IMAGE_OBJS := $(addprefix $(OBJDIR), 
$(patsubst %.cc,%.o,$(patsubst %.c,%.o,$(LABEL_IMAGE_SRCS))))

# What sources we want to compile, must be kept in sync with the main Bazel
# build files.

PROFILER_SRCS := 
    tensorflow/contrib/lite/profiling/time.cc
PROFILE_SUMMARIZER_SRCS := 
    tensorflow/contrib/lite/profiling/profile_summarizer.cc 
    tensorflow/core/util/stats_calculator.cc

CORE_CC_ALL_SRCS := 
$(wildcard tensorflow/contrib/lite/*.cc) 
$(wildcard tensorflow/contrib/lite/*.c)
ifneq ($(BUILD_TYPE),micro)
CORE_CC_ALL_SRCS += 
$(wildcard tensorflow/contrib/lite/kernels/*.cc) 
$(wildcard tensorflow/contrib/lite/kernels/internal/*.cc) 
$(wildcard tensorflow/contrib/lite/kernels/internal/optimized/*.cc) 
$(wildcard tensorflow/contrib/lite/kernels/internal/reference/*.cc) 
$(PROFILER_SRCS) 
$(wildcard tensorflow/contrib/lite/kernels/*.c) 
$(wildcard tensorflow/contrib/lite/kernels/internal/*.c) 
$(wildcard tensorflow/contrib/lite/kernels/internal/optimized/*.c) 
$(wildcard tensorflow/contrib/lite/kernels/internal/reference/*.c) 
$(wildcard tensorflow/contrib/lite/downloads/farmhash/src/farmhash.cc) 
$(wildcard tensorflow/contrib/lite/downloads/fft2d/fftsg.c)
endif
# Remove any duplicates.
CORE_CC_ALL_SRCS := $(sort $(CORE_CC_ALL_SRCS))
CORE_CC_EXCLUDE_SRCS := 
$(wildcard tensorflow/contrib/lite/*test.cc) 
$(wildcard tensorflow/contrib/lite/*/*test.cc) 
$(wildcard tensorflow/contrib/lite/*/*/*test.cc) 
$(wildcard tensorflow/contrib/lite/*/*/*/*test.cc) 
$(wildcard tensorflow/contrib/lite/kernels/test_util.cc) 
$(MINIMAL_SRCS) 
$(LABEL_IMAGE_SRCS)
ifeq ($(BUILD_TYPE),micro)
CORE_CC_EXCLUDE_SRCS += 
tensorflow/contrib/lite/model.cc 
tensorflow/contrib/lite/nnapi_delegate.cc
endif
# Filter out all the excluded files.
TF_LITE_CC_SRCS := $(filter-out $(CORE_CC_EXCLUDE_SRCS), $(CORE_CC_ALL_SRCS))
# File names of the intermediate files target compilation generates.
TF_LITE_CC_OBJS := $(addprefix $(OBJDIR), 
$(patsubst %.cc,%.o,$(patsubst %.c,%.o,$(TF_LITE_CC_SRCS))))
LIB_OBJS := $(TF_LITE_CC_OBJS)

# Benchmark sources
BENCHMARK_SRCS_DIR := tensorflow/contrib/lite/tools/benchmark
BENCHMARK_ALL_SRCS := $(TFLITE_CC_SRCS) 
    $(wildcard $(BENCHMARK_SRCS_DIR)/*.cc) 
    $(PROFILE_SUMMARIZER_SRCS)

BENCHMARK_SRCS := $(filter-out 
    $(wildcard $(BENCHMARK_SRCS_DIR)/*_test.cc), 
    $(BENCHMARK_ALL_SRCS))

BENCHMARK_OBJS := $(addprefix $(OBJDIR), 
$(patsubst %.cc,%.o,$(patsubst %.c,%.o,$(BENCHMARK_SRCS))))

# For normal manually-created TensorFlow C++ source files.
$(OBJDIR)%.o: %.cc
    @mkdir -p $(dir $@)
    $(CXX) $(CXXFLAGS) $(INCLUDES) -c $< -o $@
# For normal manually-created TensorFlow C++ source files.
$(OBJDIR)%.o: %.c
    @mkdir -p $(dir $@)
    $(CC) $(CCFLAGS) $(INCLUDES) -c $< -o $@

# The target that's compiled if there's no command-line arguments.
all: $(LIB_PATH)  $(MINIMAL_PATH) $(LABEL_IMAGE_PATH) $(BENCHMARK_BINARY) 

# The target that's compiled for micro-controllers
micro: $(LIB_PATH)

# Gathers together all the objects we've compiled into a single '.a' archive.
$(LIB_PATH): $(LIB_OBJS)
    @mkdir -p $(dir $@)
    $(AR) $(ARFLAGS) $(LIB_PATH) $(LIB_OBJS)

$(MINIMAL_PATH): $(MINIMAL_OBJS) $(LIB_PATH)
    @mkdir -p $(dir $@)
    $(CXX) $(CXXFLAGS) $(INCLUDES) 
    -o $(MINIMAL_PATH) $(MINIMAL_OBJS) 
    $(LIBFLAGS) $(LIB_PATH) $(LDFLAGS) $(LIBS)

# $(LABEL_IMAGE_PATH): $(LABEL_IMAGE_OBJS) $(LIBS)

$(LABEL_IMAGE_PATH) : $(LABEL_IMAGE_OBJS) $(LIB_PATH)
    @mkdir -p $(dir $@)
    $(CXX) $(CXXFLAGS) $(INCLUDES) 
    -o $(LABEL_IMAGE_PATH) $(LABEL_IMAGE_OBJS)
    $(LIBFLAGS) $(LIB_PATH) $(LDFLAGS) $(LIBS)


$(BENCHMARK_LIB) : $(LIB_PATH) $(BENCHMARK_OBJS)
    @mkdir -p $(dir $@)
    $(AR) $(ARFLAGS) $(BENCHMARK_LIB) $(LIB_OBJS) $(BENCHMARK_OBJS)

benchmark_lib: $(BENCHMARK_LIB)
$(info $(BENCHMARK_BINARY))
$(BENCHMARK_BINARY) : $(BENCHMARK_LIB)
    @mkdir -p $(dir $@)
    $(CXX) $(CXXFLAGS) $(INCLUDES) 
    -o $(BENCHMARK_BINARY) 
    $(LIBFLAGS) $(BENCHMARK_LIB) $(LDFLAGS) $(LIBS)

benchmark: $(BENCHMARK_BINARY)

# Gets rid of all generated files.
clean:
    rm -rf $(MAKEFILE_DIR)/gen

# Gets rid of target files only, leaving the host alone. Also leaves the lib
# directory untouched deliberately, so we can persist multiple architectures
# across builds for iOS and Android.
cleantarget:
    rm -rf $(OBJDIR)
    rm -rf $(BINDIR)

$(DEPDIR)/%.d: ;
.PRECIOUS: $(DEPDIR)/%.d

-include $(patsubst %,$(DEPDIR)/%.d,$(basename $(TF_CC_SRCS)))

          执行： ./tensorflow/contrib/lite/tools/make/build_imx6_lib.sh

          编译过程中可能会出现一些问题，依次解决即可！

          编译完成后，在./tensorflow/contrib/lite/tools/make/gen/rpi_arm7l/bin  目录下生成可执行文件label_img

          附上自己碰到的一些问题：

             1、a/tensorflow/contrib/lite/tools/benchmark/benchmark_params.h

diff --git a/tensorflow/contrib/lite/profiling/profile_summarizer.cc b/tensorflow/contrib/lite/profiling/profile_summarizer.cc
old mode 100644
new mode 100755
index c37a096..590cd21
--- a/tensorflow/contrib/lite/profiling/profile_summarizer.cc
+++ b/tensorflow/contrib/lite/profiling/profile_summarizer.cc
@@ -83,7 +83,8 @@ OperatorDetails GetOperatorDetails(const tflite::Interpreter& interpreter,
   OperatorDetails details;
   details.name = op_name;
   if (profiling_string) {
-    details.name += ":" + string(profiling_string);
+    //wly
+    details.name += ":" + std::string(profiling_string);
   }
   details.inputs = GetTensorNames(interpreter, inputs);
   details.outputs = GetTensorNames(interpreter, outputs);

             2、./tensorflow/contrib/lite/profiling/profile_summarizer.cc

diff --git a/tensorflow/contrib/lite/tools/benchmark/benchmark_params.h b/tensorflow/contrib/lite/tools/benchmark/benchmark_params.h
old mode 100644
new mode 100755
index 33448dd..e7f63ff
--- a/tensorflow/contrib/lite/tools/benchmark/benchmark_params.h
+++ b/tensorflow/contrib/lite/tools/benchmark/benchmark_params.h
@@ -47,7 +47,8 @@ class BenchmarkParam {
   virtual ~BenchmarkParam() {}
   BenchmarkParam(ParamType type) : type_(type) {}
 
- private:
+ //private:
+ public:        //wly
   static void AssertHasSameType(ParamType a, ParamType b);
   template <typename T>
   static ParamType GetValueType();

             3、./tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h

diff --git a/tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h b/tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h
old mode 100644
new mode 100755
index 2d40f17..43c54dc
--- a/tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h
+++ b/tensorflow/contrib/lite/kernels/internal/reference/reference_ops.h
@@ -24,8 +24,9 @@ limitations under the License.
 #include <type_traits>
 
- #include "third_party/eigen3/Eigen/Core"
+//#include "third_party/eigen3/Eigen/Core"" 
 #include "tensorflow/contrib/lite/kernels/internal/common.h"
 #include "tensorflow/contrib/lite/kernels/internal/quantization_util.h"
 #include "tensorflow/contrib/lite/kernels/internal/round.h"

             4、需要下载包：protobuf

             5、将c++11  改为c++0x

     12、在PC上测试label_image

          ./label_image -v 1 -m ./mobilenet_v1_1.0_224.tflite -i ./grace_hopper.jpg -l ./imagenet_slim_labels.txt

          报错信息： bash ./label_image: cannot execute binary file:Exec format error

          原因有两个：   

              一是GCC编译时多加了一个-C，生成了二进制文件；

                     解决方法:找到GCC编译处，去除-C选项。

              二就是编译环境不同（平台芯片不一致）导致

                     解决方法：需要在对应平台编译。

                            在tensorflow根目录执行：bazel build tensorflow/examples/label_image:label_image

                            如果是第一次编译，时间较久；

                            编译完成后，生成可执行文件：bazel_bin/tensorflow/examples/label_image/label_image

              本地测试该文件：

                      拷贝label_image和libtensorflow_framework.so到tensorflow/examples/label_image下

                      (第一次测试时，未拷贝libtensorflow_framework.so，直接提示：error while loading shared libraryies:libtensorflow_framework.so:cannot open shared object file:No such file or directory)

                      再次运行./tensorflow/examples/label_image/label_image

                       显示结果：military uniform(653):0.834306

                                         mortarboard(668):0.0218695

                                         academic gown(401):0.0103581

                                         pickelhaube(716):0.00800814

                                         bulletproof vest(466):0.00535084

                        说明测试OK！

      

以下操作在ARM板子上：

      1、拷贝生成的label_image到板子上（bazel_bin/tensorflow/contrib/lite/examples/label_image/）(1.7M)

            拷贝生成的label_image到板子上（bazel_bin/tensorflow/examples/label_image）(52.2M)

      2、拷贝图片./tensorflow/examples/label_image/data/grace_hopper.jpg到板子上  

      3、下载模型mobilenet_quant_v1_224.tflite

          （如果下载其它模型，可参考该文件：/tensorflow/contrib/lite/g3doc/models.md）

      4、下载模型所需文件：

             curl -L "https://storage.googleapis.com/download.tensorflow.org/models/inception_v3_2016_08_28_frozen.pb.tar.gz" | tar -C tensorflow/examples/label_image/data -xz

      5、拷贝libtensorflow_framework.so到板子上，文件位于目录bazel_bin/tensorflow/下（14M）

      4、运行label_image

        确保如下所需文件都已完成：

      （拷贝文件1：label_image）

      （拷贝文件2：grace_hopper.jpg）

      （拷贝文件3：mobilenet_quant_v1_224.tflite）

      （拷贝文件4：imagenet_slim_labels.txt）

      （拷贝文件5：libtensorflow_framework.so）

        ARM板上常用操作：

        mount /dev/sba1 /mnt/usb

        ls /mnt/usb

        cp /mnt/usb/   .

        执行脚本：

            ./label_image

           如果出现-sh: ./label_image: not found，可能是编译器不一致导致。

           尝试方法1：重定向:ln -s ld-linux.so.3 ld-linux-armhf.so.3

               新报错：./label_image:/lib/libm.so.6: version 'GLIBC_2.27' not found (required by ./label_image)

           出现以下信息，说明已经OK，恭喜你!