Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Missing CI for OPTIMIZED_KERNEL_DIR=cmsis_nn with MVEI extension #46829

Closed
mansnils opened this issue Feb 1, 2021 · 1 comment
Closed

Missing CI for OPTIMIZED_KERNEL_DIR=cmsis_nn with MVEI extension #46829

mansnils opened this issue Feb 1, 2021 · 1 comment
Assignees
@mansnils
Labels
comp:micro Related to TensorFlow Lite Microcontrollers type:bug Bug

Comments

@mansnils
Copy link
Contributor

mansnils commented Feb 1, 2021

@tensorflow/micro

System information

  • Host OS Platform and Distribution (e.g., Linux Ubuntu 16.04):
  • TensorFlow installed from (source or binary):
  • Tensorflow version (commit SHA if source):
  • Target platform (e.g. Arm Mbed OS, Arduino Nano 33 etc.):

Describe the problem
The CI script tensorflow/lite/micro/tools/ci_build/test_stm32f4.sh tests OPTIMIZED_KERNEL_DIR=cmsis_nn with DSP extension.
However there is no equivalent test for MVEI extension, i.e. Cortex-M55.

Please provide the exact sequence of commands/steps when you ran into the problem
@mansnils mansnils added the comp:micro Related to TensorFlow Lite Microcontrollers label Feb 1, 2021
@mansnils mansnils mentioned this issue Feb 1, 2021
Merged
@advaitjain advaitjain assigned mansnils and unassigned amahendrakar Feb 1, 2021
@amahendrakar amahendrakar added the type:bug Bug label Feb 8, 2021
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 10, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
57a6974 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

Progress towards tensorflow#46829
@advaitjain advaitjain mentioned this issue Feb 10, 2021
Merged
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 11, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
dcd8bd6 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

Progress towards tensorflow#46829
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 11, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
99b240e 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

In the past, we have added state to the DebugLog() and
GetCurrentTimeTicks() functions as a way to avoid having an
InitializeTarget function. With this change, we are deciding to go with
an explicit intitialization step instead.

This change has added calls to tflite::InitializeTarget to the tests,
benchmarks, and examples and converted the Arduino and SparkfunEdge to
make use of this explicit initialization.

The changes for the Arduino and SparkfunEdge have not been tested on
actual hardware.

Progress towards tensorflow#46829
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 11, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
f01253a 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

In the past, we have added state to the DebugLog() and
GetCurrentTimeTicks() functions as a way to avoid having an
InitializeTarget function. With this change, we are deciding to go with
an explicit intitialization step instead.

This change has added calls to tflite::InitializeTarget to the tests,
benchmarks, and examples and converted the Arduino and SparkfunEdge to
make use of this explicit initialization.

The changes for the Arduino and SparkfunEdge have not been tested on
actual hardware.

Progress towards tensorflow#46829
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 11, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
a8cc378 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

In the past, we have added state to the DebugLog() and
GetCurrentTimeTicks() functions as a way to avoid having an
InitializeTarget function. With this change, we are deciding to go with
an explicit intitialization step instead.

This change has added calls to tflite::InitializeTarget to the tests,
benchmarks, and examples and converted the Arduino and SparkfunEdge to
make use of this explicit initialization.

The changes for the Arduino and SparkfunEdge have not been tested on
actual hardware.

Progress towards tensorflow#46829
advaitjain added a commit to advaitjain/tensorflow that referenced this issue Feb 12, 2021
@advaitjain
Add an InitializeTarget function that can be sepcialized for a given …
81d33f1 
…target.

This will allow the unit tests to be run on additional targets that need
some addiitonal initialization (for example cornstone_300 from tensorflow#46830).

This particular change is broken out from the Cornstone PR tensorflow#46830 to
be able to have smaller more reviewable PRs.

In the past, we have added state to the DebugLog() and
GetCurrentTimeTicks() functions as a way to avoid having an
InitializeTarget function. With this change, we are deciding to go with
an explicit intitialization step instead.

This change has added calls to tflite::InitializeTarget to the tests,
benchmarks, and examples and converted the Arduino and SparkfunEdge to
make use of this explicit initialization.

The changes for the Arduino and SparkfunEdge have not been tested on
actual hardware.

Progress towards tensorflow#46829
@mansnils mansnils mentioned this issue Feb 23, 2021
Merged
@mansnils mansnils closed this as completed May 7, 2021
@google-ml-butler
Copy link

Are you satisfied with the resolution of your issue?
Yes
No

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@mansnils mansnils

Labels
comp:micro Related to TensorFlow Lite Microcontrollers type:bug Bug
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@advaitjain @mansnils @amahendrakar