Age | Commit message (Collapse) | Author |
|
Updated to latest version of TensorFlow Lite Micro and
removed forcing TENSORFLOW_LITE_MICRO_TARGET_ARCH to
Arm Cortex-M55 for SSE-310 (now it is correctly set to
Cortex-M85 instead).
Change-Id: Ib1e8680a8d2ea803e0e003032119a10b1d71754e
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Documentation updates around use of Arm® Corstone™-310 AVH, especially
in terms of use of timing adapters (TA). Also, use of TA's is forced
off for SSE-310 subsystem's CMake configuration.
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
Change-Id: I66a87060d8d47ce2580aa15f3908be20162eab54
|
|
Documents current differences between timing adapters implementations on
Corstone-300 and Corstone-310 platforms.
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
Change-Id: I3161dc929bd01217a4992be869f13377a58e5471
|
|
Assuming cmake 3.21 adds basic configuration and build presets:
* Target mps3 board with gcc
* Target mps3 board with armclang
* Target simple platform with gcc
* Target simple platform with armclang
* Target native build system
Adds test preset for native build to run default ctests.
Adds respective doc section to building instructions.
Note: hidden configurationPresets are to be moved to respective project
directories once cmake 3.24 is supported in the codebase and preset
include is supported.
Note: it is encouraged to create a personal CMakeUserPresets.json with
personal preferences like specific `LOG_LEVEL` and `jobs` for faster
compilation time.
Usage example:
cmake --preset=mps3-gcc
Signed-off-by: Maksims Svecovs <maksims.svecovs@arm.com>
Change-Id: I07a8861ed91160cc3f5167e16be45a921b60a285
|
|
Change-Id: Idfee5b33b31cabdf7e825c2161df1e71191d115a
|
|
Change-Id: I3e69514aac22973f5c2395371f1311bd5bbb83a3
|
|
Signed-off-by: Eanna O Cathain <eanna.ocathain@arm.com>
Change-Id: Ica67662d20b0c02b75418f79d4ba2f4d18373310
|
|
Added a Dockerfile to setup and install requirements for
the ml-embedded-evaluation-kit repository.
Also included minor change (MLCE-859) for NPU components'
source file.
Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: I621fb2991218a451a0e8c33e14032b9c9e36709e
|
|
Update Vela to 3.4.0
Signed-off-by: Richard Burton <richard.burton@arm.com>
Change-Id: If203d29d6884219cd4947a5753a1fb8d8284f174
|
|
Documentation updated to reflect refactoring changes
in the last few weeks.
Change-Id: Ic7abf5cc3af9898123049e890c189ed74e505309
|
|
* update to Arm Virtual Hardware link
Change-Id: If93d20e32b32ff4d6bcfa8f0e0ef096f3fff9492
|
|
* CMake version requirement bumped up to 3.21.0
* CMake 3.22.4 installed in the local Python virtualenv
* CPU flags updated in toolchain files.
* Using __ARM_FEATURE_DSP instead of potentially defining
ARM_MATH_DSP wrongly.
* CMake project version bumped up to 22.05.0
Changes also made for MLECO-3107 (pack generation):
* TensorFlow Lite Micro CMSIS-pack version updated to
1.22.02.
* Change to using __ARM_FEATURE_DSP will also help the
generated pack.
Partial changes for MLECO-3095:
* CMSIS updated to version post 5.9.0
* TensorFlow Lite Micro updated to latest available
* Ethos-U driver and core-platform repositories updated
to 20.05_rc2 tags.
Change-Id: I012c9e65897aed8ce589cff9bfe3a19efc3edeb9
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Change-Id: Ifa4b11154478355c10cb3e747b9938a74afd242b
Signed-off-by: Eanna O Cathain <eanna.ocathain@arm.com>
|
|
Certain UCs required additional work due to case context variables which also
became part of a namespace in generated files.
Solution was to declare these extra variables as part of the UC namespace in the respective model.hpp files.
Additional changes to standardise use of namespaces may be required - proposing new task.
Minor typo and rewording of customizing.md in relevant sections included.
Signed-off-by: Liam Barry <liam.barry@arm.com>
Change-Id: Ie78f82a30be252cb841136ea5115f21fc8d762cb
|
|
Change-Id: If5cf20e189273a887917aefcc745c4992be93e1d
Reviewed-on: https://eu-gerrit-2.euhpc.arm.com/c/ml/ecosystem/ml-embedded-evaluation-kit/+/543186
Tested-by: mlecosys <mlecosys@arm.com>
Reviewed-by: Nina Drozd <nina.drozd@arm.com>
Reviewed-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Platform agnostic application sources are moved into application
api module with their own independent CMake projects.
Changes for MLECO-3080 also included - they create CMake projects
individial API's (again, platform agnostic) that dependent on the
common logic. The API for KWS_API "joint" API has been removed and
now the use case relies on individual KWS, and ASR API libraries.
Change-Id: I1f7748dc767abb3904634a04e0991b74ac7b756d
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Cleaning up HAL sources by removing unnecessary redirections
with function pointers. The "platform packages" under HAL are
now streamlined enough to not need any major HAL wrapping (as
was the case before).
This allows us to have a very thin HAL layer that sits on top
of the platform and compnent packs. Also helps in getting rid
of "hal platform" pointer being passed around in the code to
use any HAL functionality.
Change-Id: I04b2057f972aad7a5cfb4a396bcdf147c9f9ef1c
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Further to the HAL refactoring done in previous commits,
this CR simpifies HAL by removing data_acq and data_psn
"modules". The associated function pointers have been
removed.
Change-Id: I04c194c08dfe0aff98ce4e0f0f056bac254c137d
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Attempting to have timer functionality contained within
the platform drivers "package" as it should (in line with
the refactoring work done so far under HAL). This will
ensure that we don't need two timer implementations under
HAL "profiles" and therefore, this whole directory can be
removed.
This change also addressed issue with the applicatio level
Profiler code knowing about how the PMU has been set up by
the platform code. This link has been removed completely.
This will make it much easier to add/amend the Ethos-U PMU
event counters types and give each platform the capability
of populating their relvant counters. The application level
Profiler doesn't know which metrics it is displaying but
just calculates and maintains statistics for whatever PMU
counters it receives from the HAL level.
A fix for timing adapter issue introduced in the last CR
is also included.
Change-Id: Ia46e03a06e7b8e42b9ed2ba8f2af2dcd2229c110
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Moving away from CMake description of targets and generation of
platform header files (for memory addresses and IRQ numbers).
Instead these headers are part of the repository under their
respective platform-driver packages under HAL sources.
Change-Id: I9bd3e68eb17385f8b93eb3d8d76b212ce0e1a6d5
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Dcoumentation updates in line with recent refactoring and some
minor update to the HAL's NPU component.
Change-Id: Iadb34dbcdedf7259f786c42bd8fcf2d950a51410
Signed-off-by: Kshitij Sisodia <kshitij.sisodia@arm.com>
|
|
Corrections and improvements in the build graph (documentation).
Change-Id: Ic7f9afe587978f835a88109c21a4145b91daf45d
|
|
Update to the build graph image in documentation for changes
made in commit c22e80e (MLECO-2919).
Change-Id: I16a5c4b11979d13010454d2f827e4ed3ca6cf5c3
|
|
Ubuntu 18.04 (or older) users
Change-Id: If7bb02f44f8bb7b84b7b7a620e69429a7b5fbb60
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
Change-Id: I771089443a6cdb4135570fcdd6e321a8f68fcf4a
|
|
* LCD module component created (removed from individual
platform packs).
* retarget.c moved out into its own component that wraps
the uart module. It also have the native stub for
GetLine => paved the way for removing data_acq module
from profiles.
* shortened names for components' dir for npu and ta
* remove peripheral_memmap and peripheral_irqs headers
from platform_drivers.h. There should be no need for
these to be included in the top level now. These should
be private headers.
* cmsis_device moved in as a component.
* Pyenv created by set_up_default_resource.py will also
install packages that CMake's source generator needs.
TODO's:
* Remove timer from profiles (MLECO-3096)
Change-Id: I9d6ea2f4f291788f40a16ed507019563c8d7f205
|
|
Change-Id: I24cd544780f46fcec8f154b440f7bb959c20a459
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
Tried to clarify location of axf file and need to change directory at a specific step
Added note on changing permissions to AVH guide and added link to AVH guide in
quickstart.md as well as recommending a VNC viewer if using AVH
Change-Id: I653f0ac3ae52a6fc1c44e28024b9513af89c6db6
Signed-off-by: Liam Barry <liam.barry@arm.com>
|
|
Change-Id: I958aa912d135241743c95a4ece81bb6deee22006
|
|
* Use UART component from Ethos-U core-platform
* Components as cmake project
Change-Id: I053f3a59800abc20216e6a4df8555db357e3f21d
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
* Core-platform now a depdenency for timing adapter
Signed-off-by: Richard Burton <richard.burton@arm.com>
Change-Id: I54f3773bdf91083fa1513be15372fcf2231bea2c
|
|
MLECO-2949: Platform drivers should own NPU and TA init
Change-Id: I13606a0197f137816bae803eb9d7d46c358b5fb8
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
Change-Id: I2036332cffee94f1a3b69b391586d0a2f6c9db55
|
|
guidance to add custom platform.
Signed-off-by: alexander <alexander.efremov@arm.com>
Change-Id: Ib2eb2b7460c0ee8161403e5b135cd8b5cd854334
|
|
* TensorFlow CMake uses Arm Ethos-U NPU driver lib path
* armclang DWARF level config flag removed
* uses and mentions of DWARF level removed
* TensorFlow uses Arm Cortex-M generic target
* removed Arm Cortex-M custom target makefile and dir
Change-Id: Ie0d3edb5473462ef8552f8b9f832190de5e99a7c
|
|
Signed-off-by: Michael Levit michaell@emza-vs.com
Change-Id: If59b6bc63375c3f1172505684d75097ff526e32e
(cherry picked from commit fba4e3a12b50d0ebb3804c6bf22cf688fcdaeee8)
|
|
* Adding troubleshooting for cloning errors
Change-Id: Ib57353900064c68c30ecb2141602081180c50c99
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
Change-Id: Ic14e93a50fb7b3f3cfd9497bac1280794cc0fc15
Signed-off-by: Isabella Gottardi <isabella.gottardi@arm.com>
|
|
Signed-off-by: Michael Levit michaell@emza-vs.com
Change-Id: I7958b05b5dbe9a785e0f8a241b716c17a9ca976f
|
|
Added SoftMax function to Mathutils to allow MicroNet
to output probability as it does not nativelu have this layer.
Minor refactoring to accommodate Softmax Calculations
Extensive renaming and updating of documentation and resource download script.
Added SoftMax function to Mathutils to allow MicroNet
to output probability.
Change-Id: I7cbbda1024d14b85c9ac1beea7ca8fbffd0b6eb5
Signed-off-by: Liam Barry <liam.barry@arm.com>
|
|
Change-Id: I2b2685dd65e08e3d8b52d223cfad53e90a73d2ba
|
|
-Fixed noise_reduction CMake bug cause inference to fail when non-default mac_units passed to build script
-Inference complete message added to noise_reduction LCD
-Doc update clarfiy vela.ini config file for compiling to non-default memory modes
-QOL improvement by adding a note on how to make path variables permenant when configuring Arm compilers
Change-Id: Id798b25638260721d8e48468b7a5942bd802d63b
|
|
Arena-cache-size is different from default when using
Dedicated_Sram memory mode.
Change-Id: Ie112146218e1ec456e17babd4ed3e7c7bc2009a8
|
|
Signed-off-by: Richard Burton <richard.burton@arm.com>
Change-Id: I47879089734cd3ab70ef8068277e677742e1b2b3
|
|
* Documentation added with missing prerequisite
* build_default.py can now be tuned for constraint build systems
Change-Id: I74c061359ff663335e664528c4f0616f55cff0f7
|
|
These changes will limit the use of FPGA internal SRAM from a max
of 4MiB to 2MiB and the BRAM from 2MiB to 1MiB.
Change-Id: I69c8e695aee26ff4f235bfe83ffd26efbd66f547
|
|
For easier look up and maintenance, all common CMake user options
have been consolidated in one CMake file.
NOTE: the individual use case specific options are still within
the correspoinding use case CMake files.
Change-Id: Id887f7b2c763f4d3eb997d997cf466684d0089b6
|
|
Change-Id: I0ea57b69bdb095f2580d80001d145fefed5cd72c
Signed-off-by: George Gekov <george.gekov@arm.com>
|
|
-Updated CMake to use C++ 14
-Removed cmsis makefile include from tensorflow.cmake
-Documentation update
-Added more flags for processor in CMake
Change-Id: I1c2b72141e98a5cf8bb09176d7c331da3b05b4c5
|
|
Change-Id: I3ec18583c321eb2815a670d56f4958e610331d6d
|