1 # ONNX operators that the Arm NN SDK supports
3 This reference guide provides a list of ONNX operators the Arm NN SDK currently supports.
5 The Arm NN SDK ONNX parser currently only supports fp32 operators.
11 See the ONNX [Add documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Add) for more information
15 See the ONNX [AveragePool documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#AveragePool) for more information.
19 See the ONNX [Constant documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Constant) for more information.
23 See the ONNX [GlobalAveragePool documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#GlobalAveragePool) for more information.
27 See the ONNX [max_pool documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#MaxPool) for more information.
31 See the ONNX [Relu documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Relu) for more information.
35 See the ONNX [Reshape documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Reshape) for more information.
37 ## Partially supported
41 The parser only supports 2D convolutions with a dilation rate of [1, 1] and group = 1 or group = #Nb_of_channel (depthwise convolution)
42 See the ONNX [Conv documentation](https://github.com/onnx/onnx/blob/master/docs/Operators.md#Conv) for more information.
44 **BatchNormalization**
46 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.
50 The parser only supports constant weights in a fully connected layer.
54 Arm tested these operators with the following ONNX fp32 neural networks:
56 * Simple MNIST. See the ONNX [MNIST documentation](https://github.com/onnx/models/tree/master/mnist) for more information.
58 * Mobilenet_v2. See the ONNX [MobileNet documentation](https://github.com/onnx/models/tree/master/models/image_classification/mobilenet) for more information.
60 More machine learning operators will be supported in future releases.