onnx_graphsurgeon.importers.onnx_importer Namespace Reference

Classes
class	OnnxImporter

Functions
List[int]	get_onnx_tensor_shape (Union[onnx.ValueInfoProto, onnx.TensorProto] onnx_tensor)
np.dtype	get_onnx_tensor_dtype (Union[onnx.ValueInfoProto, onnx.TensorProto] onnx_tensor)
Graph	import_onnx ("onnx.ModelProto" onnx_model)

Variables
	ATTR_TYPE_MAPPING = dict(zip(onnx.AttributeProto.AttributeType.values(), onnx.AttributeProto.AttributeType.keys()))
dictionary	ONNX_PYTHON_ATTR_MAPPING

Function Documentation

get_onnx_tensor_shape()

List[int] onnx_graphsurgeon.importers.onnx_importer.get_onnx_tensor_shape

(

Union[onnx.ValueInfoProto, onnx.TensorProto]

onnx_tensor

)

Here is the caller graph for this function:

get_onnx_tensor_dtype()

np.dtype onnx_graphsurgeon.importers.onnx_importer.get_onnx_tensor_dtype

(

Union[onnx.ValueInfoProto, onnx.TensorProto]

onnx_tensor

)

Here is the caller graph for this function:

import_onnx()

Graph onnx_graphsurgeon.importers.onnx_importer.import_onnx

(

"onnx.ModelProto"

onnx_model

)

Import an onnx-graphsurgeon Graph from the provided ONNX model.

Args:
    onnx_model (onnx.ModelProto): The ONNX model.

Returns:
    Graph: A corresponding onnx-graphsurgeon Graph.

Variable Documentation

ATTR_TYPE_MAPPING

onnx_graphsurgeon.importers.onnx_importer.ATTR_TYPE_MAPPING = dict(zip(onnx.AttributeProto.AttributeType.values(), onnx.AttributeProto.AttributeType.keys()))

ONNX_PYTHON_ATTR_MAPPING

dictionary onnx_graphsurgeon.importers.onnx_importer.ONNX_PYTHON_ATTR_MAPPING

Initial value:

=  {
    "FLOAT": "f",
    "INT": "i",
    "STRING": "s",
    "TENSOR": "t",
    "GRAPH": "g",
    "FLOATS": "floats",
    "INTS": "ints",
    "STRINGS": "strings",
}