mirror of
https://github.com/clearlinux/graphene.git
synced 2026-08-28 21:35:52 +00:00
Michał Kowalczyk
fcf0a01d7d
We don't use it in tests, plus it didn't work on SGX - there was a warning about autogeneration inserted before the autogenerated shebang. Also, test_106_manifest_with_shebang didn't actually test the shebang but ran the binary through the loader, so it was meaningless. We'll need to fix it and implement again after cleaning up Graphene invocation.
This example demonstrates how to run TensorFlow (v1.9) Lite's label_image example on Graphene. Reads an input image image.bmp from the current directory and uses TensorFlow lite and the Inception v3 model to label the image.
Known limitations:
- Tested on Ubuntu 16.04 with Graphene commit 030a088. Ubuntu 18.04 should work, but have not tested.
To install build dependencies on Ubuntu 16.04 there is a convenience target invoked with make install-dependencies-ubuntu. This also serves as a starting point to figure out which packages to install on newer releases of Ubuntu.
To build TensorFlow and Graphene artifacts:
- without SGX do
make - with SGX do
make SGX=1
To run the image labeling example of TensorFlow:
- without Graphene do
make run-native - with Graphene do
make run-graphene - with Graphene-SGX do
make SGX=1 run-graphene