.xoptim.patterns_ml¶

modules

.xoptim.patterns_ml.tree_ensemble

experimental_experiment.xoptim.patterns_ml.get_ml_patterns(verbose: int = 0) → List[PatternOptimization][source]¶

Returns a default list of optimization patterns for ai.onnx.ml. It is equal to the following list.

<<<

from experimental_experiment.xoptim.patterns_api import pattern_table_doc
from experimental_experiment.xoptim.patterns_ml import (
    get_ml_patterns,
)

print(pattern_table_doc(get_ml_patterns(), as_rst=True))

>>>

	name	short_name	priority	doc
0	TreeEnsembleRegressorConcatPattern	TreeEnsembleRegressorConcat	1	Replaces multiple TreeEnsembleRegressor + Concat(., axis=1) with one TreeEnsembleRegressor. All trees must have only one target (it can be extended to multiple) and is assigned a distinct dimension. The aggregation must be SUM.
1	TreeEnsembleRegressorMulPattern	TreeEnsembleRegressorMul	1	Replaces TreeEnsembleRegressor + Mul(., scalar) with TreeEnsembleRegressor.