rm multitask from MTS

examples/mts_conformal_not_sims.py

@@ -20,8 +20,8 @@
 print(198*0.8)
 # df_train = df.iloc[0:97,]
 # df_test = df.iloc[97:123,]
-df_train = df.iloc[0:158,]
-df_test = df.iloc[158:198,]
+df_train = df.iloc[0:158,:]
+df_test = df.iloc[158:198,:]
 
 regr = ns.PredictionInterval(obj=Ridge(),
                              method="splitconformal",
@@ -33,6 +33,8 @@
 obj_MTS = ns.MTS(regr, lags = 25, n_hidden_features=10, verbose = 1)
 obj_MTS.fit(df_train)
 print("\n")
+print(obj_MTS.fit_objs_)
+print("\n")
 print(obj_MTS.predict(h=10, return_pi=True))
 
 
@@ -98,5 +100,5 @@
                 lags=25)
         regr.fit(df_train)
         print(regr.predict(h=10, return_pi=True))
-      except:
+      except Exception as e:
         pass

nnetsauce.egg-info/PKG-INFO

@@ -0,0 +1,25 @@
+Metadata-Version: 2.1
+Name: nnetsauce
+Version: 0.29.2
+Summary: Quasi-randomized (neural) networks
+Home-page: https://techtonique.github.io/nnetsauce/
+Download-URL: https://github.com/Techtonique/nnetsauce
+Author: T. Moudiki
+Author-email: thierry.moudiki@gmail.com
+License: BSD Clause Clear
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+License-File: LICENSE
+Requires-Dist: joblib
+Requires-Dist: matplotlib
+Requires-Dist: numpy
+Requires-Dist: pandas
+Requires-Dist: requests
+Requires-Dist: scipy
+Requires-Dist: scikit-learn
+Requires-Dist: statsmodels
+Requires-Dist: threadpoolctl
+Requires-Dist: tqdm
+
+Quasi-randomized (neural) networks for regression, classification and time series forecasting

nnetsauce.egg-info/SOURCES.txt

@@ -0,0 +1,92 @@
+LICENSE
+README.md
+setup.py
+nnetsauce/__init__.py
+nnetsauce.egg-info/PKG-INFO
+nnetsauce.egg-info/SOURCES.txt
+nnetsauce.egg-info/dependency_links.txt
+nnetsauce.egg-info/requires.txt
+nnetsauce.egg-info/top_level.txt
+nnetsauce/base/__init__.py
+nnetsauce/base/base.py
+nnetsauce/base/baseRegressor.py
+nnetsauce/boosting/__init__.py
+nnetsauce/boosting/adaBoostClassifier.py
+nnetsauce/boosting/bst.py
+nnetsauce/custom/__init__.py
+nnetsauce/custom/custom.py
+nnetsauce/custom/customClassifier.py
+nnetsauce/custom/customRegressor.py
+nnetsauce/datasets/__init__.py
+nnetsauce/datasets/dowload.py
+nnetsauce/deep/__init__.py
+nnetsauce/deep/deepClassifier.py
+nnetsauce/deep/deepMTS.py
+nnetsauce/deep/deepRegressor.py
+nnetsauce/glm/__init__.py
+nnetsauce/glm/glm.py
+nnetsauce/glm/glmClassifier.py
+nnetsauce/glm/glmRegressor.py
+nnetsauce/lazypredict/__init__.py
+nnetsauce/lazypredict/config.py
+nnetsauce/lazypredict/lazydeepClassifier.py
+nnetsauce/lazypredict/lazydeepMTS.py
+nnetsauce/lazypredict/lazydeepRegressor.py
+nnetsauce/mts/__init__.py
+nnetsauce/mts/classical.py
+nnetsauce/mts/mts.py
+nnetsauce/multitask/__init__.py
+nnetsauce/multitask/multitaskClassifier.py
+nnetsauce/multitask/simplemultitaskClassifier.py
+nnetsauce/nonconformist/__init__.py
+nnetsauce/nonconformist/acp.py
+nnetsauce/nonconformist/base.py
+nnetsauce/nonconformist/cp.py
+nnetsauce/nonconformist/evaluation.py
+nnetsauce/nonconformist/icp.py
+nnetsauce/nonconformist/nc.py
+nnetsauce/nonconformist/util.py
+nnetsauce/optimizers/__init__.py
+nnetsauce/optimizers/helpers.py
+nnetsauce/optimizers/optimizer.py
+nnetsauce/predictioninterval/__init__.py
+nnetsauce/predictioninterval/predictioninterval.py
+nnetsauce/predictionset/__init__.py
+nnetsauce/predictionset/predictionset.py
+nnetsauce/randombag/__init__.py
+nnetsauce/randombag/bag.py
+nnetsauce/randombag/helpers.py
+nnetsauce/randombag/randomBagClassifier.py
+nnetsauce/randombag/randomBagRegressor.py
+nnetsauce/ridge2/__init__.py
+nnetsauce/ridge2/ridge2.py
+nnetsauce/ridge2/ridge2Classifier.py
+nnetsauce/ridge2/ridge2MultitaskClassifier.py
+nnetsauce/ridge2/ridge2Regressor.py
+nnetsauce/rvfl/__init__.py
+nnetsauce/rvfl/bayesianrvfl2Regressor.py
+nnetsauce/rvfl/bayesianrvflRegressor.py
+nnetsauce/sampling/__init__.py
+nnetsauce/sampling/copulas.py
+nnetsauce/sampling/helpers.py
+nnetsauce/sampling/rowsubsampling.py
+nnetsauce/simulation/__init__.py
+nnetsauce/simulation/getsims.py
+nnetsauce/simulation/nodesimulation.py
+nnetsauce/simulation/sobol.py
+nnetsauce/updater/__init__.py
+nnetsauce/updater/classification_updater.py
+nnetsauce/updater/regression_updater.py
+nnetsauce/utils/__init__.py
+nnetsauce/utils/activations.py
+nnetsauce/utils/lmfuncs.py
+nnetsauce/utils/matrixops.py
+nnetsauce/utils/memoize.py
+nnetsauce/utils/misc.py
+nnetsauce/utils/model_selection.py
+nnetsauce/utils/progress_bar.py
+nnetsauce/utils/psdcheck.py
+nnetsauce/utils/timeseries.py
+nnetsauce/utils/where.py
+nnetsauce/votingregressor/__init__.py
+nnetsauce/votingregressor/votingregressor.py

nnetsauce.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

nnetsauce.egg-info/requires.txt

@@ -0,0 +1,10 @@
+joblib
+matplotlib
+numpy
+pandas
+requests
+scipy
+scikit-learn
+statsmodels
+threadpoolctl
+tqdm

nnetsauce.egg-info/top_level.txt

@@ -0,0 +1 @@
+nnetsauce

nnetsauce/mts/mts.py

@@ -444,33 +444,33 @@ def fit(self, X, xreg=None, **kwargs):
             "block-bootstrap",
         ) or self.type_pi.startswith("vine"):
 
-            try:  # multioutput regressor
-
-                self.y_means_ = np.mean(self.y_, axis=0)
-                centered_y = self.y_ - self.y_means_
-                self.obj.fit(X=scaled_Z, y=centered_y)
-                residuals_ = centered_y - self.obj.predict(scaled_Z)
-                self.residuals_ = residuals_
-
-            except Exception:  # single output regressor
-
-                if self.show_progress is True:
-                    iterator = tqdm(range(p))
-                else:
-                    iterator = range(p)
-
-                for i in iterator:
-                    y_mean = np.mean(self.y_[:, i])
-                    self.y_means_[i] = y_mean
-                    centered_y_i = self.y_[:, i] - y_mean
-                    self.centered_y_is_.append(centered_y_i)
-                    self.obj.fit(X=scaled_Z, y=centered_y_i)
-                    self.fit_objs_[i] = deepcopy(self.obj)
-                    residuals_.append(
-                        (
-                            centered_y_i - self.fit_objs_[i].predict(scaled_Z)
-                        ).tolist()
-                    )
+            # try:  # multioutput regressor
+
+            #     self.y_means_ = np.mean(self.y_, axis=0)
+            #     centered_y = self.y_ - self.y_means_
+            #     self.obj.fit(X=scaled_Z, y=centered_y)
+            #     residuals_ = centered_y - self.obj.predict(scaled_Z)
+            #     self.residuals_ = residuals_
+
+            #except Exception as e:  # single output regressor
+            #    print(e)
+            if self.show_progress is True:
+                iterator = tqdm(range(p))
+            else:
+                iterator = range(p)
+
+            for i in iterator:
+                y_mean = np.mean(self.y_[:, i])
+                self.y_means_[i] = y_mean
+                centered_y_i = self.y_[:, i] - y_mean
+                self.centered_y_is_.append(centered_y_i)
+                self.obj.fit(X=scaled_Z, y=centered_y_i)
+                self.fit_objs_[i] = deepcopy(self.obj)
+                residuals_.append(
+                    (
+                        centered_y_i - self.fit_objs_[i].predict(scaled_Z)
+                    ).tolist()
+                )
 
         if self.type_pi.startswith("scp"):
 
@@ -791,23 +791,25 @@ def predict(self, h=5, level=95, **kwargs):
                     )
 
             if "return_pi" in kwargs:
-                try:
-                    preds_pi = self.obj.predict(cooked_new_X, return_pi=True)
+                if self.n_series <= 1: 
+                    preds_pi = self.obj.predict(cooked_new_X, 
+                                                return_pi=True)
                     mean_pi_.append(preds_pi.mean[0])
                     lower_pi_.append(preds_pi.lower[0])
                     upper_pi_.append(preds_pi.upper[0])
-                except Exception:
+                else:
                     for i in range(self.n_series):
                         preds_pi = self.fit_objs_[i].predict(
-                            cooked_new_X, return_pi=True
+                            cooked_new_X, return_pi=True,                             
                         )
                         mean_pi_.append(preds_pi.mean[0])
                         lower_pi_.append(preds_pi.lower[0])
                         upper_pi_.append(preds_pi.upper[0])
 
             try:
                 predicted_cooked_new_X = self.obj.predict(cooked_new_X)
-            except Exception:
+            except Exception as e:
+                print(e)
                 predicted_cooked_new_X = np.asarray(
                     [
                         np.asarray(

setup.py

@@ -3,7 +3,7 @@
 from codecs import open
 from os import path
 
-__version__ = '0.29.1'
+__version__ = '0.29.2'
 
 # get the dependencies and installs
 here = path.abspath(path.dirname(__file__))