Måns Nilsson comments

Results 18 comments of


                                            Måns Nilsson

CMSIS-NN : Determining output offset, input offset and output activation min and max for arm_convolve_s8()

Hi @satyapreetsingh, Cortex-R is not officially supported in TFLM but you could start with https://github.com/tensorflow/tflite-micro/tree/main/tensorflow/lite/micro/cortex_m_generic And add a case for R4 here https://github.com/tensorflow/tflite-micro/blob/main/tensorflow/lite/micro/tools/make/targets/cortex_m_generic_makefile.inc For example else ifeq ($(TARGET_ARCH), cortex-r4) CORE=R4...

CMSIS-NN basic_test for test_arm_convsolve_s8.c

Hi @jiaodawuyanzu To clarify the basic_input[40] and the PregeneratedData are exactly the same values (-9 == -9.000000000000000000e+00 etc). It is done so for convenience: https://github.com/ARM-software/CMSIS_5/blob/1cdd0e824db7340b543a7806e9a73244e73c675e/CMSIS/NN/Tests/UnitTest/generate_test_data.py#L77 And the offset corresponds to...

Integrating my tflite model with CMSIS-NN library to run on cortex-A7 processor

Hi @supratimc239 No CMSIS-NN can not in itself handle an input model in .tflite format. For that you should use it together with TFLM. Here is a starting point: https://github.com/tensorflow/tflite-micro/blob/main/tensorflow/lite/micro/kernels/cmsis_nn/README.md....

Deploying a 2D Convolutional Network from Keras

Maybe this is related to that for fully connected per layer quantization is used however for conv Keras is using per channel quantization and the legacy API is using per...

Building only CMSIS-NN Library

@uu-praveeng I am not sure what you mean by changelist? Could you try the toolchain-cmake-file from Arm Ethos-U Core Platform? For example: git clone "https://review.mlplatform.org/ml/ethos-u/ethos-u-core-platform" cmake -DCMAKE_TOOLCHAIN_FILE="X:\path\to\cloned\project\core_platform\cmake\toolchain\arm-none-eabi-gcc.cmake" -DTARGET_CPU="cortex-m3" -G "Unix...

Conflict with bazel scripts and installed tflite_micro wheel

Summarizing the problem, taking [generate_micro_mutable_op_resolver_from_model.py](https://github.com/tensorflow/tflite-micro/blob/7dbd9d23734c8c36ddf8e15b1d1c8a3346f5264c/tensorflow/lite/micro/tools/gen_micro_mutable_op_resolver/generate_micro_mutable_op_resolver_from_model.py#L26) as example. When running the script as regular python script without bazel, it does not work because visualize is not part of the tflite-micro wheel....

Måns Nilsson