-
Notifications
You must be signed in to change notification settings - Fork 74k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
TensorFlow Lite Evaluation Tools CMake Compilation #53691
Conversation
@terryheo Could you help to review CMake related changes? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for your contribution.
tensorflow/lite/tools/cmake/native_tools/protobuf/CMakeLists.txt
Outdated
Show resolved
Hide resolved
57e0343
to
ee01333
Compare
@terryheo thanks for the review, I applied the necessary fixes and squashed the commits. |
@terryheo one outstanding conflict has just been resolved - any other update required? Thanks! |
@terryheo Can you please review this PR ? Thank you! |
'-DTFLITE_KERNEL_TEST=on' flag. Unit test cross-compilation specifics can be | ||
found in the next subsection. | ||
In order to build kernel tests and/or evaluation tools, you need to provide | ||
`-DTFLITE_KERNEL_TEST=on`, `-DTFLITE_EVAL_TOOLS=on` flags, respectively. Cross-compilation specifics |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think it's better to build like benchmark_model and larbel_image without using TFLITE_KERNEL_TEST
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hello @terryheo, you meant TFLITE_EVAL_TOOLS, not TFLITE_KERNEL_TEST, right? The latter is already a part of the master branch. Under this assumption I refactored the Evaluation Tools in the most recent commit to be built similarly to benchmark_model and label_image. Please let me know if there is anything else.
de04501
to
3765e43
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@madaosik Can you please resolve conflicts? Thank you! |
@madaosik Can you please resolve conflicts? Thank you! |
1 similar comment
@madaosik Can you please resolve conflicts? Thank you! |
@madaosik Can you please resolve conflicts? Thank you! |
Hi @madaosik I'm going to go ahead and close this PR, because it seems to have stalled. If you're still interested in pursing this (and responding to my comments), please feel free to reopen! Thank you! |
This PR introduces into CMake configuration of TF Lite the option to automatically (cross-)compile TF Lite evaluation tools (specifically ImageNet Image Classification, COCO Object Detection and Indifference Diff).
Similarly to TF Lite kernel tests cross-compilation requiring host
flatc
compiler, cross-compilation of evaluation tools requires hostprotoc
compiler. Therefore, the CMake configuration for obtainingflatc
has been extended and abstracted into the native_tools CMake module.For details please see the updated
build_cmake.md
file.