diff options
| author | Conor Kennedy <conor.kennedy@arm.com> | 2019-05-13 14:48:58 +0100 |
|---|---|---|
| committer | conor.kennedy <conor.kennedy@arm.com> | 2019-05-14 09:44:13 +0000 |
| commit | 305620201159eb47e005a46d60f80c6a2c763d51 (patch) | |
| tree | 9d42e71f24c37f94b7094e14489cd6ae459fc786 | |
| parent | eac246dde1c494ffdffbcf68930141c32e244ad1 (diff) | |
| download | armnn-305620201159eb47e005a46d60f80c6a2c763d51.tar.gz | |
IVGCVSW-3069 Add documentation for ModelAccuracyTool-Armnn
Signed-off-by: Conor Kennedy <conor.kennedy@arm.com>
Change-Id: I1e128982ba0e1c99ee4532276f679dda37095201
| -rw-r--r-- | tests/ModelAccuracyTool-Armnn/ModelAccuracyTool-Armnn.cpp | 2 | ||||
| -rw-r--r-- | tests/ModelAccuracyTool-Armnn/README.md | 22 |
2 files changed, 23 insertions, 1 deletions
diff --git a/tests/ModelAccuracyTool-Armnn/ModelAccuracyTool-Armnn.cpp b/tests/ModelAccuracyTool-Armnn/ModelAccuracyTool-Armnn.cpp index 7b968302d9..3abfb3c2ec 100644 --- a/tests/ModelAccuracyTool-Armnn/ModelAccuracyTool-Armnn.cpp +++ b/tests/ModelAccuracyTool-Armnn/ModelAccuracyTool-Armnn.cpp @@ -106,7 +106,7 @@ int main(int argc, char* argv[]) { // Adds generic options needed to run Accuracy Tool. desc.add_options() - ("help", "Display help messages") + ("help,h", "Display help messages") ("model-path,m", po::value<std::string>(&modelPath)->required(), "Path to armnn format model file") ("compute,c", po::value<std::vector<armnn::BackendId>>(&computeDevice)->default_value(defaultBackends), backendsMessage.c_str()) diff --git a/tests/ModelAccuracyTool-Armnn/README.md b/tests/ModelAccuracyTool-Armnn/README.md new file mode 100644 index 0000000000..2a8c977968 --- /dev/null +++ b/tests/ModelAccuracyTool-Armnn/README.md @@ -0,0 +1,22 @@ +# The ModelAccuracyTool-Armnn + +The `ModelAccuracyTool-Armnn` is a program for measuring the Top 5 accuracy results of a model against an image dataset. + +Prerequisites: +1. The model is in .armnn format model file. The `ArmnnConverter` can be used to convert a model to this format. +2. The ImageNet test data is in raw tensor file format. The `ImageTensorGenerator` can be used to convert the test +images to this format. + +|Cmd:||| +| ---|---|---| +| -h | --help | Display help messages | +| -m | --model-path | Path to armnn format model file | +| -c | --compute | Which device to run layers on by default. Possible choices: CpuRef, CpuAcc, GpuAcc | +| -d | --data-dir | Path to directory containing the ImageNet test data | +| -i | --input-name | Identifier of the input tensors in the network separated by comma | +| -o | --output-name | Identifier of the output tensors in the network separated by comma | +| -v | --validation-labels-path | Path to ImageNet Validation Label file | + +Example usage: <br> +<code>./ModelAccuracyTool -m /path/to/model/model.armnn -c CpuRef -d /path/to/test/directory/ -i input -o output +-v /path/to/file/val.txt</code> |