nnoir-onnx 1.0.4

ONNX to NNOIR Converter

nnoir-onnx

nnoir-onnx is a converter from ONNX model to NNOIR model.

Install

pip install nnoir-onnx

Example

wget https://www.cntk.ai/OnnxModels/mnist/opset_7/mnist.tar.gz
tar xvzf mnist.tar.gz
onnx2nnoir -o model.nnoir mnist/model.onnx

Supported ONNX Operators

• Add
• AveragePool
• BatchNormalization
• Clip
  • min must be 0
• Concat
• Conv
  • W must be Constant value or have initializer value
  • b must be Constant value or have initializer value
• Dropout
  • equivalent identity function
• Elu
• Flatten
• Gemm
  • B must be Constant value or have initializer value
  • C must be Constant value or have initializer value
• GlobalAveragePool
• LRN
• MatMul
  • B must be Constant value or have initializer value
• MaxPool
• Pad
  • mode must be "constant"
• ReduceSum
• Relu
• Reshape
• Resize
  • must be from opset version >= 11
  • mode must be "linear"
  • coordinate_transformation_mode must be either "pytorch_half_pixel" or "align_corners"
• Sigmoid
• Softmax
• Sub
  • 1st input must not be "constant"
• Sum
  • 2 inputs
• Tanh
• Transpose