Support all architectures using 'org.tensorflow:tensorflow-android' dependency #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Otherwise Android Studio shows warnings for every log line
This will work on API levels less than 18
I’ve copied the approach used by the Android Things sample app: https://github.com/androidthings/sample-tensorflow-imageclassifier This dependency downloads an AAR file which includes native binaries for all architectures, along with a Java interface (TensorFlowInferenceInterface) than can be used to interface with TensorFlow model without needing to write any JNI code! Documentation: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/android TESTING Before these changes, I could not install the app on my Android 7.0 x86 emulator. It failed with this message: `Failure [INSTALL_FAILED_NO_MATCHING_ABIS: Failed to extract native libraries, res=-113]` After the change, I was successfully able to launch the app and Detect digits. :-)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
The
org.tensorflow:tensorflow-android:1.2.0dependency downloads an AAR file which includes native binaries for all architectures, along with a Java interface (TensorFlowInferenceInterface) than can be used to interface with TensorFlow model, all without needing to write any JNI code!Documentation:
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/contrib/android
I've written this blog post providing a bit more background information:
https://medium.com/@daj/using-a-pre-trained-tensorflow-model-on-android-e747831a3d6
This PR also includes commits to remove IDE warnings due to:
TESTING
Before these changes, I could not install the app on my Android 7.0 x86 emulator. It failed with this message:
Failure [INSTALL_FAILED_NO_MATCHING_ABIS: Failed to extract native libraries, res=-113]After the change, I was successfully able to launch the app and Detect digits. :-)