1# Unit tests for CMSIS-NN 2 3Unit test CMSIS-NN functions on any [Arm Mbed OS](https://os.mbed.com/mbed-os/) supported HW or using a fixed virtual platform (FVP) based on [Arm Corstone-300 software](https://developer.arm.com/ip-products/subsystem/corstone/corstone-300). 4 5The [Unity test framework](http://www.throwtheswitch.org/unity) is used for running the actual unit tests. 6 7## Requirements 8 9Python3 is required. 10It has been tested with Python 3.6 and it has been tested on Ubuntu 16, 18 and 20. 11 12Make sure to use the latest pip version before starting. 13If in a virtual environment just start by upgrading pip. 14 15``` 16pip install --upgrade pip 17``` 18 19See below for what pip packages are needed. 20 21### Executing unit tests 22 23If using the script unittest_targets.py for executing unit tests, the following packages are needed. 24 25``` 26pip install pyserial mbed-ls termcolor 27``` 28 29Other required python packages are mbed-cli and and mbed-ls. It should not matter if those are installed under python2 or python3 as they are command-line tools. These packages have been tested for Python2, with the following versions: mbed-ls(1.7.9) and mbed-cli(1.10.1). 30 31### Generating new test data 32 33For generating new test data, the following packages are needed. 34 35``` 36pip install numpy packaging tensorflow 37``` 38 39 40For generating new data, the python3 packages tensorflow, numpy and packaging are required. Most unit tests use a Keras generated model for reference. The SVDF unit test use a json template as input for generating a model. To do so flatc compiler is needed and it requires a schema file. 41 42#### Get flatc and schema 43 44Note this is only needed for generating SVDF unit tests. 45 46For flatc compiler clone this [repo](https://github.com/google/flatbuffers) and build: 47``` 48cd flatbuffers 49cmake -G "Unix Makefiles" -DCMAKE_BUILD_TYPE=Release 50make 51``` 52Remember to add the built flatc binary to the path. 53 54For schema file download [schema.fbs](https://raw.githubusercontent.com/tensorflow/tensorflow/master/tensorflow/lite/schema/schema.fbs). 55 56#### Using tflite_runtime 57Python package tensorflow is always needed however the script has the option to use tflite_runtime for the interpreter, which will generate the actual reference output. Python package tflite_runtime can be installed with pip and it can also be built locally. Check this [link](https://www.tensorflow.org/lite/guide/build_cmake_pip) on how to do that. 58To use the tflite_runtime the script currently has to be modified. 59 60## Getting started 61 62 63 64### Using Arm Mbed OS supported hardware 65 66Connect any HW (e.g. NUCLEO_F746ZG) that is supported by Arm Mbed OS. Multiple boards are supported. If all requirements are satisfied you can just run: 67 68``` 69./unittest_targets.py 70``` 71 72Use the -h flag to get more info. 73 74### Using FVP based on Arm Corstone-300 software 75 76The build for unit tests differs from the build of CMSIS-NN as a [standalone library](https://github.com/ARM-software/CMSIS-NN/blob/main/README.md#building-cmsis-nn-as-a-library) in that, there is a dependency to [CMSIS](https://github.com/ARM-software/CMSIS_5) project for the startup files from CMSIS-Core. This is specified by the mandatory CMSIS_PATH CMake argument. 77 78 79Here is an example for testing on Arm Cortex-M55: 80 81``` 82cd </path/to/CMSIS_NN/Tests/Unittest> 83mkdir build 84cd build 85cmake .. -DCMAKE_TOOLCHAIN_FILE=</path/to/ethos-u-core-platform>/cmake/toolchain/arm-none-eabi-gcc.cmake -DTARGET_CPU=cortex-m55 -DCMSIS_PATH=</path/to/CMSIS> 86make 87``` 88 89This will build all unit tests. You can also just build a specific unit test only, for example: 90 91``` 92make test_arm_depthwise_conv_s8_opt 93``` 94 95Then you need to download and install the FVP based Arm Corstone-300 software, for example: 96 97``` 98mkdir -p /home/$user/FVP 99wget https://developer.arm.com/-/media/Arm%20Developer%20Community/Downloads/OSS/FVP/Corstone-300/FVP_Corstone_SSE-300_Ethos-U55_11.12_57.tgz 100tar -xvzf FVP_Corstone_SSE-300_Ethos-U55_11.12_57.tgz 101./FVP_Corstone_SSE-300_Ethos-U55.sh --i-agree-to-the-contained-eula --no-interactive -d /home/$user/FVP 102export PATH="/home/$user/FVP/models/Linux64_GCC-6.4:$PATH" 103``` 104 105Finally you can run the unit tests. For example: 106 107``` 108FVP_Corstone_SSE-300_Ethos-U55 --cpulimit 2 -C mps3_board.visualisation.disable-visualisation=1 -C mps3_board.telnetterminal0.start_telnet=0 -C mps3_board.uart0.out_file="-" -C mps3_board.uart0.unbuffered_output=1 ./TestCases/test_arm_depthwise_conv_s8_opt/test_arm_depthwise_conv_s8_opt.elf 109``` 110 111## Generating new test data 112 113Generating new test data is done with the following script. Use the -h flag to get more info. 114 115``` 116./generate_test_data.py -h 117 118``` 119 120The script use a concept of test data sets, i.e. it need a test set data name as input. It will then generate files with that name as prefix. Multiple header files of different test sets can then be included in the actual unit test files. 121When adding a new test data set, new c files should be added or existing c files should be updated to use the new data set. See overview of the folders on how/where to add new c files. 122 123The steps to add a new unit test are as follows. Add a new test test in the load_all_testdatasets() function. Run the generate script with that new test set as input. Add the new generated header files to an existing or new unit test. 124 125### Tests depending on specific TFL versions or patched TFL version 126 127#### LSTM 128 129The LSTM tests are using the tflite_runtime as interpreter. 130See [Using tflite_runtime](https://github.com/ARM-software/CMSIS-NN/blob/main/Tests/UnitTest/README.md#using-tflite_runtime) for more info. 131This patch is needed for the tflite_runtime (or tensorflow if using that): 132https://github.com/tensorflow/tflite-micro/pull/1253 - Note that this PR is for [TFLM](https://github.com/tensorflow/tflite-micro) so it has to be ported to [TFL](https://github.com/tensorflow/tensorflow) before building the tflite_runtime. 133The issue related to this is: https://github.com/tensorflow/tflite-micro/issues/1455 134 135 136## Overview of the Folders 137 138- `Corstone-300` - These are dependencies, like linker files etc, needed when building binaries targeting the FVP based on Arm Corstone-300 software. This is mostly taken from Arm Ethos-U Core Platform project. 139- `Mbed` - These are the Arm Mbed OS settings that are used. See Mbed/README.md. 140- `Output` - This will be created when building. 141- `PregeneratedData` - Host local(Not part of GitHub) test data for model creation using Keras in unit tests. It can be used for debug purposes when 142 adding new operators or debugging existing ones. 143- `TestCases` - Here are the actual unit tests. For each function under test there is a folder under here. 144- `TestCases/<cmsis-nn function name>` - For each function under test there is a folder with the same name with test_ prepended to the name and it contains a c-file with the actual unit tests. For example for arm_convolve_s8() the file is called test_arm_convolve_s8.c 145- `TestCases/<cmsis-nn function name>/Unity` - This folder contains a Unity file that calls the actual unit tests. For example for arm_convolve_s8() the file is called unity_test_arm_convolve_s8.c. 146- `TestCases/<cmsis-nn function name>/Unity/TestRunner` - This folder will contain the autogenerated Unity test runner. 147- `TestCases/TestData` - This is auto generated test data in .h files that the unit tests are using. The data in PregenrateData folder has fp32 data of a network whereas here it is the quantized equivalent of the same. They are not the same. All data can regenerated or only parts of it (e.g. only bias data). Of course even the config can be regenerated. This partial/full regeneration is useful during debugging. 148- `TestCases/Common` - Common files used in test data generation is placed here. 149 150## Formatting 151 152The python test scripts should be formatted with yapf. 153 154``` 155pip install --upgrade yapf 156``` 157 158The following settings are used. 159 160``` 161python -m yapf --in-place --style='{based_on_style:pep8,column_limit:120,indent_width:4}' *.py 162```