From d786531442fdbaba164853b87a4f8b6c700d6748 Mon Sep 17 00:00:00 2001
From: Luc Giffon <luc.giffon@lis-lab.fr>
Date: Mon, 4 Nov 2019 17:03:30 +0100
Subject: [PATCH] now the model can make predictions: todo: manage result
recording
---
code/bolsonaro/models/omp_forest_regressor.py | 50 ++++++++++++++++---
code/bolsonaro/trainer.py | 1 +
code/train.py | 5 +-
3 files changed, 48 insertions(+), 8 deletions(-)
diff --git a/code/bolsonaro/models/omp_forest_regressor.py b/code/bolsonaro/models/omp_forest_regressor.py
index 7813da9..2f87892 100644
--- a/code/bolsonaro/models/omp_forest_regressor.py
+++ b/code/bolsonaro/models/omp_forest_regressor.py
@@ -34,6 +34,9 @@ class OmpForestRegressor(BaseEstimator):
def models_parameters(self):
return self._models_parameters
+ def score_regressor(self, X, y):
+ return self._regressor.score(X, y)
+
def _train_forest(self, X_train, y_train):
self._regressor.fit(X_train, y_train)
forest = self._regressor.estimators_
@@ -51,24 +54,57 @@ class OmpForestRegressor(BaseEstimator):
:return:
"""
self._logger.debug("Forest make prediction on X_train")
- D = np.array([tree.predict(X_train) for tree in self._forest]).T
+ D = self._forest_prediction(X_train)
if self._models_parameters.normalize:
self._logger.debug("Compute norm of predicted vectors on X_train")
self._forest_norms = np.linalg.norm(D, axis=0)
D /= self._forest_norms
-
omp = OrthogonalMatchingPursuit(
n_nonzero_coefs=self._models_parameters.extracted_forest_size,
fit_intercept=False, normalize=False)
self._logger.debug("Apply orthogonal maching pursuit on forest for {} extracted trees."
.format(self._models_parameters.extracted_forest_size))
omp.fit(D, y_train)
- weights = omp.coef_ # why not to use directly the omp estimator and bypass it using the coefs?
+ weights = omp.coef_
+ # question: why not to use directly the omp estimator instead of bypassing it using the coefs?
return weights
- def predict(self):
- raise NotImplementedError("TODO: implement predict function")
- # todo don't forget to deal with the normalize parameter
- # should the norm used on train or the new norms be used for normalization?
+ def _forest_prediction(self, X):
+ return np.array([tree.predict(X) for tree in self._forest]).T
+
+ def predict(self, X):
+ """
+ Apply the OMPForestRegressor to X.
+
+ :param X:
+ :return:
+ """
+ D = self._forest_prediction(X)
+
+ if self._models_parameters.normalize:
+ D /= self._forest_norms
+
+ predictions = D @ self.weights
+
+ return predictions
+
+
+ def score(self, X, y, metric="mse"):
+ """
+ Evaluate OMPForestRegressor on (`X`, `y`) using `metric`
+
+ :param X:
+ :param y:
+ :param metric:
+ :return:
+ """
+ predictions = self.predict(X)
+
+ if metric == "mse":
+ evaluation = np.mean(np.square(predictions - y))
+ else:
+ raise ValueError("Metric value {} is not known.")
+
+ return evaluation
\ No newline at end of file
diff --git a/code/bolsonaro/trainer.py b/code/bolsonaro/trainer.py
index 95fab4a..0e239e6 100644
--- a/code/bolsonaro/trainer.py
+++ b/code/bolsonaro/trainer.py
@@ -17,6 +17,7 @@ class Trainer(object):
# why is this function named iterate?
self._logger.info('Training model using train set...')
begin_time = time.time()
+ # todo: OMP may be running with X_dev ou Y_dev
model.fit(self._dataset.X_train, self._dataset.y_train)
end_time = time.time()
diff --git a/code/train.py b/code/train.py
index 9a50283..7b589a3 100644
--- a/code/train.py
+++ b/code/train.py
@@ -20,7 +20,7 @@ if __name__ == "__main__":
load_dotenv(find_dotenv())
default_dataset_name = 'boston'
- default_normalize = False
+ default_normalize = True
default_forest_size = 100
default_extracted_forest_size = 10
# the models will be stored in a directory structure like: models/{experiment_id}/seeds/{seed_nb}/extracted_forest_size/{nb_extracted_trees}
@@ -100,3 +100,6 @@ if __name__ == "__main__":
model = ModelFactory.build(dataset.task, model_parameters)
trainer.iterate(model, sub_models_dir)
+
+ print(model.score(dataset.X_test, dataset.y_test))
+ print(model.score_regressor(dataset.X_test, dataset.y_test))
\ No newline at end of file
--
GitLab