TensorFlow Lite is TensorFlow's lightweight solution for Objective-C developers. It enables low-latency inference of on-device machine learning models with a small binary size and fast performance supporting hardware acceleration.
To build the Objective-C TensorFlow Lite library on Apple platforms,
install from source
or clone the GitHub repo.
Then, configure TensorFlow by navigating to the root directory and executing the
configure.py
script:
python configure.py
Follow the prompts and when asked to build TensorFlow with iOS support, enter y
.
Add the TensorFlow Lite pod to your Podfile
:
pod 'TensorFlowLiteObjC'
Then, run pod install
.
In your Objective-C files, import the umbrella header:
#import "TFLTensorFlowLite.h"
Or, the module if you set CLANG_ENABLE_MODULES = YES
in your Xcode project:
@import TFLTensorFlowLite;
Note: To import the TensorFlow Lite module in your Objective-C files, you must
also include use_frameworks!
in your Podfile
.
In your BUILD
file, add the TensorFlowLite
dependency to your target:
objc_library(
deps=[
"//tensorflow/lite/objc:TensorFlowLite",
],)
In your Objective-C files, import the umbrella header:
#import "TFLTensorFlowLite.h"
Or, the module if you set CLANG_ENABLE_MODULES = YES
in your Xcode project:
@import TFLTensorFlowLite;
Build the TensorFlowLite
Objective-C library target:
bazel build tensorflow/lite/objc:TensorFlowLite
Build the tests
target:
bazel test tensorflow/lite/objc:tests
Open the //tensorflow/lite/objc/TensorFlowLite.tulsiproj
using
the TulsiApp
or by running the
generate_xcodeproj.sh
script from the root tensorflow
directory:
generate_xcodeproj.sh --genconfig tensorflow/lite/objc/TensorFlowLite.tulsiproj:TensorFlowLite --outputfolder ~/path/to/generated/TensorFlowLite.xcodeproj