From 19e1cebcc6e385288c5104cd303a8a8225d24a47 Mon Sep 17 00:00:00 2001
From: Danilo Ferreira de Lima <danilo.enoque.ferreira.de.lima@xfel.de>
Date: Sat, 6 May 2023 12:33:19 +0200
Subject: [PATCH] Removed multi output optimization for ARD, since it does not
work in this case. Use generic parallelization.
---
pes_to_spec/model.py | 59 +++++++++++++++++++++++++++++++++++++++++---
1 file changed, 56 insertions(+), 3 deletions(-)
diff --git a/pes_to_spec/model.py b/pes_to_spec/model.py
index 10df31f..2ff7ab4 100644
--- a/pes_to_spec/model.py
+++ b/pes_to_spec/model.py
@@ -362,7 +362,7 @@ def _fit_estimator(estimator, X: np.ndarray, y: np.ndarray, w: Optional[np.ndarr
estimator.fit(X, y, w)
return estimator
-class MultiOutputWithStd(MetaEstimatorMixin, BaseEstimator):
+class MultiOutputRidgeWithStd(MetaEstimatorMixin, BaseEstimator):
def __init__(self, estimator, *, n_jobs=-1):
self.estimator = estimator
@@ -416,6 +416,59 @@ class MultiOutputWithStd(MetaEstimatorMixin, BaseEstimator):
y_std = np.sqrt(sigmas_squared_data + self.fast_inv_alpha)
return y, y_std
+class MultiOutputGenericWithStd(MetaEstimatorMixin, BaseEstimator):
+
+ def __init__(self, estimator, *, n_jobs=-1):
+ self.estimator = estimator
+ self.n_jobs = n_jobs
+
+ def fit(self, X: np.ndarray, y: np.ndarray, weights: Optional[np.ndarray]=None):
+ """Fit the model to data, separately for each output variable.
+
+ Args:
+ X: {array-like, sparse matrix} of shape (n_samples, n_features)
+ The input data.
+ y: {array-like, sparse matrix} of shape (n_samples, n_outputs)
+ Multi-output targets. An indicator matrix turns on multilabel
+ estimation.
+
+ Returns: self.
+ """
+ if y.ndim == 1:
+ raise ValueError(
+ "y must have at least two dimensions for "
+ "multi-output regression but has only one."
+ )
+ self.estimators_ = Parallel(n_jobs=self.n_jobs)(
+ delayed(_fit_estimator)(
+ self.estimator, X, y[:, i], weights
+ )
+ for i in range(y.shape[1])
+ )
+
+ return self
+
+ def predict(self, X: np.ndarray, return_std: bool=False) -> Union[np.ndarray, Tuple[np.ndarray, np.ndarray]]:
+ """Predict multi-output variable using model for each target variable.
+
+ Args:
+ X: {array-like, sparse matrix} of shape (n_samples, n_features)
+ The input data.
+
+ Returns: {array-like, sparse matrix} of shape (n_samples, n_outputs)
+ Multi-output targets predicted across multiple predictors.
+ Note: Separate models are generated for each predictor.
+ """
+ y = Parallel(n_jobs=self.n_jobs)(
+ delayed(e.predict)(X, return_std) for e in self.estimators_
+ )
+ if return_std:
+ y, unc = zip(*y)
+ return np.asarray(y).T, np.asarray(unc).T
+
+ return np.asarray(y).T
+
+
class UncorrelatedDeviation(OutlierMixin, BaseEstimator):
"""
Detect outliers from uncorrelated inputs.
@@ -550,9 +603,9 @@ class Model(TransformerMixin, BaseEstimator):
if model_type == "bnn":
self.fit_model = BNNModel()
elif model_type == "ridge":
- self.fit_model = MultiOutputWithStd(BayesianRidge(n_iter=300, verbose=True), n_jobs=8)
+ self.fit_model = MultiOutputRidgeWithStd(BayesianRidge(n_iter=300, verbose=True), n_jobs=8)
elif model_type == "ard":
- self.fit_model = MultiOutputWithStd(ARDRegression(n_iter=300, verbose=True), n_jobs=8)
+ self.fit_model = MultiOutputGenericWithStd(ARDRegression(n_iter=300, verbose=True), n_jobs=8)
self.model_type = model_type
self.kde_xgm = None
--
GitLab