總覽
當部署用於裝置端機器學習 (ODML) 應用程式的模型時,務必注意行動裝置上可用的記憶體有限。模型二進位檔案大小與模型中使用的運算數量密切相關。TensorFlow Lite 可讓您透過選擇性建構來縮減模型二進位檔案大小。選擇性建構會略過模型集中未使用的運算,並產生精簡的程式庫,其中僅包含在行動裝置上執行模型所需的執行階段和運算核心。
選擇性建構適用於下列三個運算程式庫。
下表說明選擇性建構對一些常見用例的影響
模型名稱 | 領域 | 目標架構 | AAR 檔案大小 |
---|---|---|---|
Mobilenet_1.0_224(float) | 圖片分類 | armeabi-v7a | tensorflow-lite.aar (296,635 位元組) |
arm64-v8a | tensorflow-lite.aar (382,892 位元組) | ||
SPICE | 聲音音調擷取 | armeabi-v7a | tensorflow-lite.aar (375,813 位元組) tensorflow-lite-select-tf-ops.aar (1,676,380 位元組) |
arm64-v8a | tensorflow-lite.aar (421,826 位元組) tensorflow-lite-select-tf-ops.aar (2,298,630 位元組) |
||
i3d-kinetics-400 | 影片分類 | armeabi-v7a | tensorflow-lite.aar (240,085 位元組) tensorflow-lite-select-tf-ops.aar (1,708,597 位元組) |
arm64-v8a | tensorflow-lite.aar (273,713 位元組) tensorflow-lite-select-tf-ops.aar (2,339,697 位元組) |
使用 Bazel 選擇性建構 TensorFlow Lite
本節假設您已下載 TensorFlow 原始碼,並設定 Bazel 的本機開發環境。
為 Android 專案建構 AAR 檔案
您可以透過提供模型檔案路徑,如下所示建構自訂 TensorFlow Lite AAR。
sh tensorflow/lite/tools/build_aar.sh \
--input_models=/a/b/model_one.tflite,/c/d/model_two.tflite \
--target_archs=x86,x86_64,arm64-v8a,armeabi-v7a
上述指令會為 TensorFlow Lite 內建和自訂運算產生 AAR 檔案 bazel-bin/tmp/tensorflow-lite.aar
;如果您的模型包含選取 TensorFlow 運算,則會選擇性地產生 aar 檔案 bazel-bin/tmp/tensorflow-lite-select-tf-ops.aar
。請注意,這會建構包含多種不同架構的「fat」AAR;如果您不需要所有架構,請使用適合您部署環境的子集。
使用自訂運算建構
如果您已開發具有自訂運算的 Tensorflow Lite 模型,您可以透過將以下標記新增至建構指令來建構這些模型
sh tensorflow/lite/tools/build_aar.sh \
--input_models=/a/b/model_one.tflite,/c/d/model_two.tflite \
--target_archs=x86,x86_64,arm64-v8a,armeabi-v7a \
--tflite_custom_ops_srcs=/e/f/file1.cc,/g/h/file2.h \
--tflite_custom_ops_deps=dep1,dep2
tflite_custom_ops_srcs
標記包含自訂運算的原始檔,而 tflite_custom_ops_deps
標記包含建構這些原始檔的依附元件。請注意,這些依附元件必須存在於 TensorFlow 存放區中。
進階用法:自訂 Bazel 規則
如果您的專案使用 Bazel,而且您想為一組給定的模型定義自訂 TFLite 依附元件,您可以在專案存放區中定義下列規則
僅適用於具有內建運算的模型
load(
"@org_tensorflow//tensorflow/lite:build_def.bzl",
"tflite_custom_android_library",
"tflite_custom_c_library",
"tflite_custom_cc_library",
)
# A selectively built TFLite Android library.
tflite_custom_android_library(
name = "selectively_built_android_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
# A selectively built TFLite C library.
tflite_custom_c_library(
name = "selectively_built_c_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
# A selectively built TFLite C++ library.
tflite_custom_cc_library(
name = "selectively_built_cc_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
適用於具有選取 TF 運算的模型
load(
"@org_tensorflow//tensorflow/lite/delegates/flex:build_def.bzl",
"tflite_flex_android_library",
"tflite_flex_cc_library",
)
# A Select TF ops enabled selectively built TFLite Android library.
tflite_flex_android_library(
name = "selective_built_tflite_flex_android_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
# A Select TF ops enabled selectively built TFLite C++ library.
tflite_flex_cc_library(
name = "selective_built_tflite_flex_cc_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
進階用法:建構自訂 C/C++ 共用程式庫
如果您想針對給定的模型建構自己的自訂 TFLite C/C++ 共用物件,您可以按照下列步驟操作
透過在 TensorFlow 原始碼的根目錄執行下列指令來建立暫時性的 BUILD 檔案
mkdir -p tmp && touch tmp/BUILD
建構自訂 C 共用物件
如果您想建構自訂 TFLite C 共用物件,請將下列程式碼新增至 tmp/BUILD
檔案
load(
"//tensorflow/lite:build_def.bzl",
"tflite_custom_c_library",
"tflite_cc_shared_object",
)
tflite_custom_c_library(
name = "selectively_built_c_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
# Generates a platform-specific shared library containing the TensorFlow Lite C
# API implementation as define in `c_api.h`. The exact output library name
# is platform dependent:
# - Linux/Android: `libtensorflowlite_c.so`
# - Mac: `libtensorflowlite_c.dylib`
# - Windows: `tensorflowlite_c.dll`
tflite_cc_shared_object(
name = "tensorflowlite_c",
linkopts = select({
"//tensorflow:ios": [
"-Wl,-exported_symbols_list,$(location //tensorflow/lite/c:exported_symbols.lds)",
],
"//tensorflow:macos": [
"-Wl,-exported_symbols_list,$(location //tensorflow/lite/c:exported_symbols.lds)",
],
"//tensorflow:windows": [],
"//conditions:default": [
"-z defs",
"-Wl,--version-script,$(location //tensorflow/lite/c:version_script.lds)",
],
}),
per_os_targets = True,
deps = [
":selectively_built_c_lib",
"//tensorflow/lite/c:exported_symbols.lds",
"//tensorflow/lite/c:version_script.lds",
],
)
新增的目標可以如下方式建構
bazel build -c opt --cxxopt=--std=c++17 \
//tmp:tensorflowlite_c
以及針對 Android (對於 64 位元,請將 android_arm
取代為 android_arm64
)
bazel build -c opt --cxxopt=--std=c++17 --config=android_arm \
//tmp:tensorflowlite_c
建構自訂 C++ 共用物件
如果您想建構自訂 TFLite C++ 共用物件,請將下列程式碼新增至 tmp/BUILD
檔案
load(
"//tensorflow/lite:build_def.bzl",
"tflite_custom_cc_library",
"tflite_cc_shared_object",
)
tflite_custom_cc_library(
name = "selectively_built_cc_lib",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
# Shared lib target for convenience, pulls in the core runtime and builtin ops.
# Note: This target is not yet finalized, and the exact set of exported (C/C++)
# APIs is subject to change. The output library name is platform dependent:
# - Linux/Android: `libtensorflowlite.so`
# - Mac: `libtensorflowlite.dylib`
# - Windows: `tensorflowlite.dll`
tflite_cc_shared_object(
name = "tensorflowlite",
# Until we have more granular symbol export for the C++ API on Windows,
# export all symbols.
features = ["windows_export_all_symbols"],
linkopts = select({
"//tensorflow:macos": [
"-Wl,-exported_symbols_list,$(location //tensorflow/lite:tflite_exported_symbols.lds)",
],
"//tensorflow:windows": [],
"//conditions:default": [
"-Wl,-z,defs",
"-Wl,--version-script,$(location //tensorflow/lite:tflite_version_script.lds)",
],
}),
per_os_targets = True,
deps = [
":selectively_built_cc_lib",
"//tensorflow/lite:tflite_exported_symbols.lds",
"//tensorflow/lite:tflite_version_script.lds",
],
)
新增的目標可以如下方式建構
bazel build -c opt --cxxopt=--std=c++17 \
//tmp:tensorflowlite
以及針對 Android (對於 64 位元,請將 android_arm
取代為 android_arm64
)
bazel build -c opt --cxxopt=--std=c++17 --config=android_arm \
//tmp:tensorflowlite
針對具有選取 TF 運算的模組,您也需要建構下列共用程式庫
load(
"@org_tensorflow//tensorflow/lite/delegates/flex:build_def.bzl",
"tflite_flex_shared_library"
)
# Shared lib target for convenience, pulls in the standard set of TensorFlow
# ops and kernels. The output library name is platform dependent:
# - Linux/Android: `libtensorflowlite_flex.so`
# - Mac: `libtensorflowlite_flex.dylib`
# - Windows: `libtensorflowlite_flex.dll`
tflite_flex_shared_library(
name = "tensorflowlite_flex",
models = [
":model_one.tflite",
":model_two.tflite",
],
)
新增的目標可以如下方式建構
bazel build -c opt --cxxopt='--std=c++17' \
--config=monolithic \
--host_crosstool_top=@bazel_tools//tools/cpp:toolchain \
//tmp:tensorflowlite_flex
以及針對 Android (對於 64 位元,請將 android_arm
取代為 android_arm64
)
bazel build -c opt --cxxopt='--std=c++17' \
--config=android_arm \
--config=monolithic \
--host_crosstool_top=@bazel_tools//tools/cpp:toolchain \
//tmp:tensorflowlite_flex
使用 Docker 選擇性建構 TensorFlow Lite
本節假設您已在本機電腦上安裝 Docker,並在此處下載 TensorFlow Lite Dockerfile 。
下載上述 Dockerfile 後,您可以執行下列程式碼來建構 docker 映像檔
docker build . -t tflite-builder -f tflite-android.Dockerfile
為 Android 專案建構 AAR 檔案
執行下列程式碼來下載使用 Docker 建構的指令碼
curl -o build_aar_with_docker.sh \
https://raw.githubusercontent.com/tensorflow/tensorflow/master/tensorflow/lite/tools/build_aar_with_docker.sh &&
chmod +x build_aar_with_docker.sh
接著,您可以透過提供模型檔案路徑,如下所示建構自訂 TensorFlow Lite AAR。
sh build_aar_with_docker.sh \
--input_models=/a/b/model_one.tflite,/c/d/model_two.tflite \
--target_archs=x86,x86_64,arm64-v8a,armeabi-v7a \
--checkpoint=master \
[--cache_dir=<path to cache directory>]
checkpoint
標記是您想要在建構程式庫之前查看的 TensorFlow 存放區的提交、分支或標籤;預設值為最新發行分支。上述指令會為 TensorFlow Lite 內建和自訂運算產生 AAR 檔案 tensorflow-lite.aar
,並選擇性地為您目前目錄中的選取 TensorFlow 運算產生 AAR 檔案 tensorflow-lite-select-tf-ops.aar
。
--cache_dir
指定快取目錄。如果未提供,指令碼會在目前工作目錄下建立名為 bazel-build-cache
的目錄以進行快取。
將 AAR 檔案新增至專案
透過將 AAR 直接匯入您的專案,或透過將自訂 AAR 發布至您的本機 Maven 存放區來新增 AAR 檔案。請注意,如果您產生了 tensorflow-lite-select-tf-ops.aar
,您也必須新增此檔案的 AAR 檔案。
iOS 的選擇性建構
請參閱「在本機建構」章節以設定建構環境並設定 TensorFlow 工作區,然後按照指南使用 iOS 的選擇性建構指令碼。