module onnx_tools.optim.onnx_optimisation#

Short summary#

module mlprodict.onnx_tools.optim.onnx_optimisation

Optimisations of ONNX graphs.

source on GitHub

Functions#

function

truncated documentation

onnx_remove_node

Removes as many nodes as possible without changing the outcome. It applies onnx_remove_node_unused(), onnx_remove_node_identity(), …

Documentation#

Optimisations of ONNX graphs.

source on GitHub

mlprodict.onnx_tools.optim.onnx_optimisation.onnx_remove_node(onnx_model, recursive=True, debug_info=None, **options)#

Removes as many nodes as possible without changing the outcome. It applies onnx_remove_node_unused, onnx_remove_node_identity, and onnx_remove_node_redundant.

Parameters:

  • onnx_model – onnx model

  • recursive – looks into subgraphs

  • debug_info – debug information (private)

  • options – additional options

Returns:

new onnx model

source on GitHub