diff options
| author | Jan Eilers <jan.eilers@arm.com> | 2021-02-19 09:38:03 +0000 |
|---|---|---|
| committer | Jan Eilers <jan.eilers@arm.com> | 2021-02-22 08:32:32 +0000 |
| commit | 1121c6bf3fc4ff34b6fce2e17a2cda71be185112 (patch) | |
| tree | 868a1cc73d7c8ce06da8d97eb1b7260647f7d472 | |
| parent | 8548e9ccb0b8a5a8780a7d35e5abf19bb13f9dbc (diff) | |
| download | armnn-1121c6bf3fc4ff34b6fce2e17a2cda71be185112.tar.gz | |
IVGCVSW-5695 Update supported operators
* Update supported operators for the delegate, parsers,
serializer and deserializer
Signed-off-by: Jan Eilers <jan.eilers@arm.com>
Change-Id: I33ac99a29d894eec055cd05411014075d78b3168
| -rw-r--r-- | README.md | 2 | ||||
| -rw-r--r-- | docs/01_01_parsers.dox | 8 | ||||
| -rw-r--r-- | docs/01_02_deserializer_serializer.dox | 4 | ||||
| -rw-r--r-- | docs/01_03_delegate.dox | 18 |
4 files changed, 25 insertions, 7 deletions
@@ -36,7 +36,7 @@ to build for a wide variety of target platforms, from a wide variety of host env Depending on what kind of framework (Tensorflow, Caffe, ONNX) you've been using to create your model there are multiple software tools available within Arm NN that can serve your needs. -Generally, there is a **parser** available **for each supported framework**. Each parser allows you to run a models from +Generally, there is a **parser** available **for each supported framework**. Each parser allows you to run models from one framework e.g. the TfLite-Parser lets you run TfLite models. You can integrate these parsers into your own application to load, optimize and execute your model. We also provide **python bindings** for our parsers and the Arm NN core. We call the result **PyArmNN**. Therefore your application can be conveniently written in either C++ using the "original" diff --git a/docs/01_01_parsers.dox b/docs/01_01_parsers.dox index 025858e13d..ae49303876 100644 --- a/docs/01_01_parsers.dox +++ b/docs/01_01_parsers.dox @@ -127,8 +127,7 @@ The Arm NN SDK ONNX parser currently only supports fp32 operators. ### Partially supported - Conv - - The parser only supports 2D convolutions with a dilation rate of [1, 1] and group = 1 or group = #Nb_of_channel (depthwise convolution) - See the ONNX [Conv documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Conv) for more information. + - The parser only supports 2D convolutions with a group = 1 or group = #Nb_of_channel (depthwise convolution) - BatchNormalization - The parser does not support training mode. See the ONNX [BatchNormalization documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#BatchNormalization) for more information. - MatMul @@ -179,13 +178,15 @@ The Arm NN SDK TensorFlow Lite parser currently supports the following operators - MAXIMUM - MEAN - MINIMUM -- MU +- MUL - NEG - PACK - PAD - QUANTIZE - RELU - RELU6 +- REDUCE_MAX +- REDUCE_MIN - RESHAPE - RESIZE_BILINEAR - RESIZE_NEAREST_NEIGHBOR @@ -197,6 +198,7 @@ The Arm NN SDK TensorFlow Lite parser currently supports the following operators - SQUEEZE - STRIDED_SLICE - SUB +- SUM - TANH - TRANSPOSE - TRANSPOSE_CONV diff --git a/docs/01_02_deserializer_serializer.dox b/docs/01_02_deserializer_serializer.dox index 047cb5dc8d..6884b93dd1 100644 --- a/docs/01_02_deserializer_serializer.dox +++ b/docs/01_02_deserializer_serializer.dox @@ -59,6 +59,7 @@ The Arm NN SDK Serializer currently supports the following layers: - Quantize - QuantizedLstm - Rank +- Reduce - Reshape - Resize - Slice @@ -143,10 +144,10 @@ The Arm NN SDK Deserialize parser currently supports the following layers: - QLstm - QuantizedLstm - Rank +- Reduce - Reshape - Resize - ResizeBilinear -- Rsqrt - Slice - Softmax - SpaceToBatchNd @@ -157,6 +158,7 @@ The Arm NN SDK Deserialize parser currently supports the following layers: - StridedSlice - Subtraction - Switch +- Transpose - TransposeConvolution2d More machine learning layers will be supported in future releases. diff --git a/docs/01_03_delegate.dox b/docs/01_03_delegate.dox index 9063f05658..f6d8e7660d 100644 --- a/docs/01_03_delegate.dox +++ b/docs/01_03_delegate.dox @@ -82,15 +82,17 @@ The Arm NN SDK TensorFlow Lite delegate currently supports the following operato - LOCAL_RESPONSE_NORMALIZATION - LOGICAL_AND -- + - LOGICAL_NOT -- + - LOGICAL_OR - LOGISTIC - LOG_SOFTMAX +- LSTM + - L2_NORMALIZATION - L2_POOL_2D @@ -111,8 +113,16 @@ The Arm NN SDK TensorFlow Lite delegate currently supports the following operato - PAD +- PRELU + - QUANTIZE +- RANK + +- REDUCE_MAX + +- REDUCE_MIN + - RESHAPE - RESIZE_BILINEAR @@ -137,8 +147,12 @@ The Arm NN SDK TensorFlow Lite delegate currently supports the following operato - SQRT +- STRIDED_SLICE + - SUB +- SUM + - TANH - TRANSPOSE |