module onnxrt.ops_onnx._op
#
Short summary#
module mlprodict.onnxrt.ops_onnx._op
Additional methods for the extension of ReferenceEvaluator.
Classes#
class |
truncated documentation |
---|---|
Base class to cache C++ implementation based on inputs. |
Properties#
property |
truncated documentation |
---|---|
|
Returns node attribute domain. |
|
Returns node attribute input. |
|
Returns node attribute op_type. |
|
Returns node attribute output. |
Methods#
method |
truncated documentation |
---|---|
Caches an implementation. |
|
Returns the cached implementation for key key. |
|
Returns a key mapped to the corresponding C++ implementation. |
|
Tells if a key belongs to the cache. |
Documentation#
Additional methods for the extension of ReferenceEvaluator.
- class mlprodict.onnxrt.ops_onnx._op.OpRunExtended(onnx_node: NodeProto, run_params: Dict[str, Any])#
Bases:
OpRun
Base class to cache C++ implementation based on inputs.
- __abstractmethods__ = frozenset({'_run'})#
- _abc_impl = <_abc._abc_data object>#
- cache_impl(key, rt)#
Caches an implementation.
- get_cache_impl(key)#
Returns the cached implementation for key key.
- get_cache_key(**kwargs)#
Returns a key mapped to the corresponding C++ implementation.
- has_cache_key(key)#
Tells if a key belongs to the cache.