module onnxrt.ops_onnxruntime._op
#
Short summary#
module mlprodict.onnxrt.ops_onnxruntime._op
Shortcut to ops_onnxruntime.
Classes#
class |
truncated documentation |
---|---|
Unique operator which calls onnxruntime to compute predictions for one operator. |
Methods#
method |
truncated documentation |
---|---|
Initializes the node. |
|
Tells the runtime if this node needs the context (all the results produced so far) as it may silently access … |
|
Should be overwritten. |
Documentation#
Shortcut to ops_onnxruntime.
- class mlprodict.onnxrt.ops_onnxruntime._op.OpRunOnnxRuntime(onnx_node, desc=None, variables=None, dtype=None, runtime=None, **options)#
Bases:
object
Unique operator which calls onnxruntime to compute predictions for one operator.
- Parameters:
onnx_node – onnx node
desc – internal representation
variables – registered variables created by previous operators
dtype – float computation type
options – runtime options
runtime – onnxruntime1, onnxruntime1-cuda, …
- __init__(onnx_node, desc=None, variables=None, dtype=None, runtime=None, **options)#
- Parameters:
onnx_node – onnx node
desc – internal representation
variables – registered variables created by previous operators
dtype – float computation type
options – runtime options
runtime – onnxruntime1, onnxruntime1-cuda, …
- _guess_proto_type(dtype)#
- _init(variables=None)#
Initializes the node.
- Parameters:
variables – registered variables created by previous operators
The current implementation for operator Scan only works for matrices.
- _name_mapping(inputs)#
- need_context()#
Tells the runtime if this node needs the context (all the results produced so far) as it may silently access one of them (operator Loop). The default answer is False.
- run(*args, **kwargs)#
Should be overwritten.