使用docker生成Tensorflow Lite aar文件

使用docker生成Tensorflow Lite aar文件

1. 下载docker镜像

下载dockerfile，重命名为tflite.Dockerfile
docker pull tensorflow/build:latest-python3.11 # 根据dockerfile里面的内容进行修改
docker build . -t tflite-builder -f tflite.Dockerfile # dockerfile中拉取某些文件，要翻墙开vpn

2. 运行docker镜像

挂载到本地目录，用于保存android库 等文件
镜像名字通过docker images查看
docker run -it --network=host -v pwd:/host_dir tflite-builder bash # linux改成 $PWD ； --network=host 是为了后面可以使用宿主机vpn

3. download additional Android tools and libraries

sdkmanager \
  "build-tools;${ANDROID_BUILD_TOOLS_VERSION}" \
  "platform-tools" \
  "platforms;android-${ANDROID_API_LEVEL}"

4. 下载tensorflow源码

cd /host_dir/
git clone https://github.com/tensorflow/tensorflow.git
git checkout 某个版本 # 如果是最新的docker镜像和代码，可以跳过

5. Configure WORKSPACE and .bazelrc

cd /host_dir/tensorflow/
./configure
配置如下，一路回车默认，interactively configure选择y

tf-docker /host_dir/tensorflow > ./configure
WARNING: current bazel installation is not a release version.
Please specify the location of python. [Default is /usr/bin/python3]:


Found possible Python library paths:
  /usr/lib/python3/dist-packages
  /usr/local/lib/python3.11/dist-packages
Please input the desired Python library path to use.  Default is [/usr/lib/python3/dist-packages]

Do you wish to build TensorFlow with ROCm support? [y/N]:
No ROCm support will be enabled for TensorFlow.

Do you wish to build TensorFlow with CUDA support? [y/N]:
No CUDA support will be enabled for TensorFlow.

Do you want to use Clang to build TensorFlow? [Y/n]:
Clang will be used to compile TensorFlow.

Please specify the path to clang executable. [Default is /usr/lib/llvm-17/bin/clang]:


You have Clang 17.0.6 installed.

Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -Wno-sign-compare]:


Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: y
Searching for NDK and SDK installations.

Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See .bazelrc for more details.
        --config=mkl            # Build with MKL support.
        --config=mkl_aarch64    # Build with oneDNN and Compute Library for the Arm Architecture (ACL).
        --config=monolithic     # Config for mostly static monolithic build.
        --config=numa           # Build with NUMA support.
        --config=dynamic_kernels        # (Experimental) Build kernels into separate shared objects.
        --config=v1             # Build with TensorFlow 1 API instead of TF 2 API.
Preconfigured Bazel build configs to DISABLE default on features:
        --config=nogcp          # Disable GCP support.
        --config=nonccl         # Disable NVIDIA NCCL support.
Configuration finished

6. Build and install

需要翻墙，设置代理
curl google.com # 测试能否翻墙

export https_proxy=http://host.docker.internal:7890;http_proxy=http://host.docker.internal:7890; all_proxy=socks5://host.docker.internal:7890
参考：
https://github.com/kubernetes-sigs/kind/issues/3416

bazel build -c opt --cxxopt=--std=c++17 --config=android_arm64 \
  --fat_apk_cpu=x86,x86_64,arm64-v8a,armeabi-v7a \
  --define=android_dexmerger_tool=d8_dexmerger \
  --define=android_incremental_dexing_tool=d8_dexbuilder \
  //tensorflow/lite/java:tensorflow-lite

ls bazel-bin/tensorflow/lite/java/*.aar
就可以看到生成的aar文件了

7. 参考

https://www.tensorflow.org/lite/android/lite_build
https://github.com/tensorflow/tensorflow/issues/60831