diff options
| -rw-r--r-- | README.md | 1 | ||||
| -rw-r--r-- | lib/model.py | 14 | ||||
| -rw-r--r-- | lib/parameters.py | 28 |
3 files changed, 38 insertions, 5 deletions
@@ -32,6 +32,7 @@ The following variables may be set to alter the behaviour of dfatool components. | `DFATOOL_DTREE_ENABLED` | 0, **1** | Use decision trees in get\_fitted | | `DFATOOL_DTREE_FUNCTION_LEAVES` | 0, **1** | Use functions (fitted via linear regression) in decision tree leaves when modeling numeric parameters with at least three distinct values. If 0, integer parameters are treated as enums instead. | | `DFATOOL_DTREE_SKLEARN_CART` | **0**, 1 | Use sklearn CART ("Decision Tree Regression") algorithm for decision tree generation. Uses binary nodes and supports splits on scalar variables. Overrides `FUNCTION_LEAVES` (=0) and `NONBINARY_NODES` (=0). | +| `DFATOOL_DTREE_SKLEARN_DECART` | **0**, 1 | Use sklearn CART ("Decision Tree Regression") algorithm for decision tree generation. Ignore scalar parameters, thus emulating the DECART algorithm. | | `DFATOOL_DTREE_LMT` | **0**, 1 | Use [Linear Model Tree](https://github.com/cerlymarco/linear-tree) algorithm for regression tree generation. Uses binary nodes and linear functions. Overrides `FUNCTION_LEAVES` (=0) and `NONBINARY_NODES` (=0). | | `DFATOOL_CART_MAX_DEPTH` | **0** .. *n* | maximum depth for sklearn CART. Default: unlimited. | | `DFATOOL_USE_XGBOOST` | **0**, 1 | Use Extreme Gradient Boosting algorithm for decision forest generation. | diff --git a/lib/model.py b/lib/model.py index 558f049..baa22da 100644 --- a/lib/model.py +++ b/lib/model.py @@ -160,6 +160,9 @@ class AnalyticModel: with_sklearn_cart = bool( int(os.getenv("DFATOOL_DTREE_SKLEARN_CART", "0")) ) + with_sklearn_decart = bool( + int(os.getenv("DFATOOL_DTREE_SKLEARN_DECART", "0")) + ) with_lmt = bool(int(os.getenv("DFATOOL_DTREE_LMT", "0"))) with_xgboost = bool(int(os.getenv("DFATOOL_USE_XGBOOST", "0"))) ignore_irrelevant_parameters = bool( @@ -178,6 +181,7 @@ class AnalyticModel: with_function_leaves=with_function_leaves, with_nonbinary_nodes=with_nonbinary_nodes, with_sklearn_cart=with_sklearn_cart, + with_sklearn_decart=with_sklearn_decart, with_lmt=with_lmt, with_xgboost=with_xgboost, ignore_irrelevant_parameters=ignore_irrelevant_parameters, @@ -332,6 +336,9 @@ class AnalyticModel: with_sklearn_cart = bool( int(os.getenv("DFATOOL_DTREE_SKLEARN_CART", "0")) ) + with_sklearn_decart = bool( + int(os.getenv("DFATOOL_DTREE_SKLEARN_DECART", "0")) + ) with_lmt = bool(int(os.getenv("DFATOOL_DTREE_LMT", "0"))) with_xgboost = bool(int(os.getenv("DFATOOL_USE_XGBOOST", "0"))) ignore_irrelevant_parameters = bool( @@ -359,6 +366,7 @@ class AnalyticModel: with_function_leaves=with_function_leaves, with_nonbinary_nodes=with_nonbinary_nodes, with_sklearn_cart=with_sklearn_cart, + with_sklearn_decart=with_sklearn_decart, with_lmt=with_lmt, with_xgboost=with_xgboost, ignore_irrelevant_parameters=ignore_irrelevant_parameters, @@ -450,6 +458,7 @@ class AnalyticModel: with_function_leaves=False, with_nonbinary_nodes=True, with_sklearn_cart=False, + with_sklearn_decart=False, with_lmt=False, with_xgboost=False, ignore_irrelevant_parameters=True, @@ -475,6 +484,7 @@ class AnalyticModel: with_function_leaves=with_function_leaves, with_nonbinary_nodes=with_nonbinary_nodes, with_sklearn_cart=with_sklearn_cart, + with_sklearn_decart=with_sklearn_decart, with_lmt=with_lmt, with_xgboost=with_xgboost, ignore_irrelevant_parameters=ignore_irrelevant_parameters, @@ -778,6 +788,9 @@ class PTAModel(AnalyticModel): with_sklearn_cart = bool( int(os.getenv("DFATOOL_DTREE_SKLEARN_CART", "0")) ) + with_sklearn_decart = bool( + int(os.getenv("DFATOOL_DTREE_SKLEARN_DECART", "0")) + ) with_lmt = bool(int(os.getenv("DFATOOL_DTREE_LMT", "0"))) with_xgboost = bool(int(os.getenv("DFATOOL_USE_XGBOOST", "0"))) ignore_irrelevant_parameters = bool( @@ -796,6 +809,7 @@ class PTAModel(AnalyticModel): with_function_leaves=with_function_leaves, with_nonbinary_nodes=with_nonbinary_nodes, with_sklearn_cart=with_sklearn_cart, + with_sklearn_decart=with_sklearn_decart, with_lmt=with_lmt, with_xgboost=with_xgboost, ignore_irrelevant_parameters=ignore_irrelevant_parameters, diff --git a/lib/parameters.py b/lib/parameters.py index bca189e..fc6512f 100644 --- a/lib/parameters.py +++ b/lib/parameters.py @@ -38,7 +38,9 @@ def distinct_param_values(param_tuples): return distinct_values -def param_to_ndarray(param_tuples, with_nan=True, categorial_to_scalar=False): +def param_to_ndarray( + param_tuples, with_nan=True, categorial_to_scalar=False, ignore_indexes=list() +): has_nan = dict() has_non_numeric = dict() distinct_values = dict() @@ -71,6 +73,9 @@ def param_to_ndarray(param_tuples, with_nan=True, categorial_to_scalar=False): else: ignore_index[i] = False + for i in ignore_indexes: + ignore_index[i] = True + ret_tuples = list() for param_tuple in param_tuples: ret_tuple = list() @@ -928,6 +933,7 @@ class ModelAttribute: with_function_leaves=False, with_nonbinary_nodes=True, with_sklearn_cart=False, + with_sklearn_decart=False, with_xgboost=False, with_lmt=False, ignore_irrelevant_parameters=True, @@ -943,6 +949,8 @@ class ModelAttribute: :param with_nonbinary_nodes: Allow non-binary nodes for enum and scalar parameters (i.e., nodes with more than two children) :param with_sklearn_cart: Use `sklearn.tree.DecisionTreeRegressor` CART implementation for tree generation. Does not support categorial (enum) and sparse parameters. Both are ignored during fitting. All other options are ignored as well. + :param with_sklearn_decart: Use `sklearn.tree.DecisionTreeRegressor` CART implementation in DECART mode for tree generation. CART limitations + apply; additionaly, scalar parameters are ignored during fitting. :param loss_ignore_scalar: Ignore scalar parameters when computing the loss for split candidates. Only sensible if with_function_leaves is enabled. :param threshold: Return a StaticFunction leaf node if std(data) < threshold. Default 100. @@ -953,16 +961,26 @@ class ModelAttribute: int(os.getenv("DFATOOL_PARAM_CATEGORIAL_TO_SCALAR", "0")) ) - if with_sklearn_cart: + if with_sklearn_cart or with_sklearn_decart: from sklearn.tree import DecisionTreeRegressor max_depth = int(os.getenv("DFATOOL_CART_MAX_DEPTH", "0")) if max_depth == 0: max_depth = None cart = DecisionTreeRegressor(max_depth=max_depth) - fit_parameters, category_to_index, ignore_index = param_to_ndarray( - parameters, with_nan=False, categorial_to_scalar=categorial_to_scalar - ) + if with_sklearn_cart: + fit_parameters, category_to_index, ignore_index = param_to_ndarray( + parameters, + with_nan=False, + categorial_to_scalar=categorial_to_scalar, + ) + elif with_sklearn_decart: + fit_parameters, category_to_index, ignore_index = param_to_ndarray( + parameters, + with_nan=False, + categorial_to_scalar=categorial_to_scalar, + ignore_indexes=self.scalar_param_indexes, + ) if fit_parameters.shape[1] == 0: logger.warning( f"Cannot generate CART for {self.name} {self.attr} due to lack of parameters: parameter shape is {np.array(parameters).shape}, fit_parameter shape is {fit_parameters.shape}" |