tensorflow_lite

TensorFlow_lite sample in Python running in Mystikos

This sample demonstrates how to invoke TensowFlow_lite to label images from within an enclave. It is written in python.

It is recommended that you understand how to use EXT2 file system and how applications can be packaged Mystkos before proceeding with this sample. Please refer to the README for the helloworld sample to obtain the necessary information.

Transform a Dockerfile into a root file system

Dockerfile.slim.buster sets up all the prerequsites to use Tensowflow. It downloads a photo, model and labels. myst-appbuilder is used to convert the Dockerfile into a directory containing all the files needed to run this application in Mystikos.

myst-appbuilder -v -d Dockerfile.slim.buster

After this, the appdir generated can be converted into a cpio archive using myst mkcpio or an EXT2 file system using myst mkext2that can be loaded into Mystikos. In this sample, we use an EXT2 file system.

After this, the appdir generated can be converted into a cpio archive using myst mkcpio or an EXT2 file system using myst mkext2that can be loaded into Mystikos. In this sample, we use an EXT2 file system. Please refer to the README for the helloworld sample for information on the EXT2 file system.

Functionality

label_image.py takes an image ( in this sample, an image of Grace Hopper is used) and runs it through a MobileNetV1 model. It then prints out the top 5 labels that match the image.

Configuration parameters

A Mystikos package needs configuration to control certain run-time environmental settings as well as settings that control how the application reads environmental data from the insecure host environment. This sample needs more memory than helloworld. ApplicationPath ensures that python3 is run after the root file system is mounted. The model, labels and image are sent in via ApplicationParameters. EnvironmentVariables is used to set up the environment variables for use inside Mystikos.

{
    "Debug": 1,
    "ProductID": 1,
    "SecurityVersion": 1,
    "UserMemSize": "2048m",
    "ApplicationPath": "/usr/local/bin/python3",
    "ApplicationParameters": ["/app/label_image.py",
        "--model_file", "/tmp/mobilenet_v1_1.0_224.tflite",
        "--label_file", "/tmp/labels.txt",
        "--image", "/tmp/grace_hopper.bmp"],
    "EnvironmentVariables": ["PYTHONPATH=/usr/lib/python3/dist-packages"]
}

To learn more about configuration, please refer to related documentation.

Running the sample

To run the sample in package mode, use make run.

To run the sample using myst exec-sgx, use make runexec. Note that the myst-exec command takes the application configuration as a parameter

@myst exec-sgx ext2rootfs /usr/local/bin/python3 --app-config-path config.json

This is because additional application configuration such as the the application parameters and environmental variables are supplied by config.json.