Fix inittime (#256)

rasbt · Sep 28, 2017 · 00981a6 · 00981a6
1 parent 3424df6
commit 00981a6
Show file tree

Hide file tree

Showing 20 changed files with 221 additions and 4 deletions.
diff --git a/docs/sources/CHANGELOG.md b/docs/sources/CHANGELOG.md
@@ -28,6 +28,7 @@ The CHANGELOG for the current development version is available at
 
 - the "S" vector from SVD in `PrincipalComponentAnalysis` are now scaled so that the eigenvalues via `solver='eigen'` and `solver='svd'` now store eigenvalues that have the same magnitudes. [#251](https://github.com/rasbt/mlxtend/pull/251)
 - The parameters for `StackingClassifier`, `StackingCVClassifier`, `StackingRegressor`, `StackingCVRegressor`, and `EnsembleVoteClassifier` can now be tuned using scikit-learn's `GridSearchCV` ([#254](https://github.com/rasbt/mlxtend/pull/254) via [James Bourbeau](https://github.com/jrbourbeau))
+- Fix issues with `self._init_time` parameter in `_IterativeModel` subclasses. [#256](https://github.com/rasbt/mlxtend/pull/256)
 
 ### Version 0.8.0 (2017-09-09)
 

diff --git a/mlxtend/_base/_base_model.py b/mlxtend/_base/_base_model.py
@@ -6,11 +6,13 @@
 #
 # License: BSD 3 clause
 
+from time import time
+
 
 class _BaseModel(object):
 
     def __init__(self):
-        pass
+        self._init_time = time()
 
     def _check_arrays(self, X, y=None):
         if isinstance(X, list):

diff --git a/mlxtend/_base/_classifier.py b/mlxtend/_base/_classifier.py
@@ -7,6 +7,7 @@
 # License: BSD 3 clause
 
 import numpy as np
+from time import time
 
 
 class _Classifier(object):
@@ -75,6 +76,7 @@ def fit(self, X, y, init_params=True):
         self._check_target_array(y)
         if hasattr(self, 'self.random_seed') and self.random_seed:
             self._rgen = np.random.RandomState(self.random_seed)
+        self._init_time = time()
         self._fit(X=X, y=y, init_params=init_params)
         self._is_fitted = True
         return self

diff --git a/mlxtend/_base/_cluster.py b/mlxtend/_base/_cluster.py
@@ -7,6 +7,7 @@
 # License: BSD 3 clause
 
 import numpy as np
+from time import time
 
 
 class _Cluster(object):
@@ -36,6 +37,7 @@ def fit(self, X, init_params=True):
         self._check_arrays(X=X)
         if hasattr(self, 'self.random_seed') and self.random_seed:
             self._rgen = np.random.RandomState(self.random_seed)
+        self._init_time = time()
         self._fit(X=X, init_params=init_params)
         self._is_fitted = True
         return self

diff --git a/mlxtend/_base/_iterative_model.py b/mlxtend/_base/_iterative_model.py
@@ -31,7 +31,7 @@ def _print_progress(self, iteration, n_iter,
                 if not hasattr(self, 'ela_str_'):
                     self.ela_str_ = '00:00:00'
                 if not iteration % time_interval:
-                    ela_sec = time() - self.init_time_
+                    ela_sec = time() - self._init_time
                     self.ela_str_ = self._to_hhmmss(ela_sec)
                 s += ' | Elapsed: %s' % self.ela_str_
                 if self.print_progress > 2:

diff --git a/mlxtend/_base/_regressor.py b/mlxtend/_base/_regressor.py
@@ -7,6 +7,7 @@
 # License: BSD 3 clause
 
 import numpy as np
+from time import time
 
 
 class _Regressor(object):
@@ -44,6 +45,7 @@ def fit(self, X, y, init_params=True):
         self._check_target_array(y)
         if hasattr(self, 'self.random_seed') and self.random_seed:
             self._rgen = np.random.RandomState(self.random_seed)
+        self._init_time = time()
         self._fit(X=X, y=y, init_params=init_params)
         self._is_fitted = True
         return self

diff --git a/mlxtend/classifier/adaline.py b/mlxtend/classifier/adaline.py
@@ -57,6 +57,10 @@ def __init__(self, eta=0.01, epochs=50,
                  minibatches=None, random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _IterativeModel.__init__(self)
+        _Classifier.__init__(self)
+
         self.eta = eta
         self.minibatches = minibatches
         self.epochs = epochs

diff --git a/mlxtend/classifier/logistic_regression.py b/mlxtend/classifier/logistic_regression.py
@@ -61,6 +61,10 @@ def __init__(self, eta=0.01, epochs=50,
                  random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _IterativeModel.__init__(self)
+        _Classifier.__init__(self)
+
         self.eta = eta
         self.epochs = epochs
         self.l2_lambda = l2_lambda

diff --git a/mlxtend/classifier/multilayerperceptron.py b/mlxtend/classifier/multilayerperceptron.py
@@ -82,6 +82,12 @@ def __init__(self, eta=0.5, epochs=50,
                  minibatches=1, random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _Classifier.__init__(self)
+        _IterativeModel.__init__(self)
+        _MultiClass.__init__(self)
+        _MultiLayer.__init__(self)
+
         if len(hidden_layers) > 1:
             raise AttributeError('Currently, only 1 hidden layer is supported')
         self.hidden_layers = hidden_layers

diff --git a/mlxtend/classifier/perceptron.py b/mlxtend/classifier/perceptron.py
@@ -49,6 +49,10 @@ class Perceptron(_BaseModel, _IterativeModel, _Classifier):
     def __init__(self, eta=0.1, epochs=50, random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _IterativeModel.__init__(self)
+        _Classifier.__init__(self)
+
         self.eta = eta
         self.epochs = epochs
         self.random_seed = random_seed

diff --git a/mlxtend/classifier/softmax_regression.py b/mlxtend/classifier/softmax_regression.py
@@ -16,7 +16,7 @@
 from .._base import _Classifier
 
 
-class SoftmaxRegression(_BaseModel, _IterativeModel, _MultiClass, _Classifier):
+class SoftmaxRegression(_BaseModel, _IterativeModel, _Classifier,  _MultiClass):
 
     """Softmax regression classifier.
 
@@ -66,6 +66,11 @@ def __init__(self, eta=0.01, epochs=50,
                  random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _IterativeModel.__init__(self)
+        _Classifier.__init__(self)
+        _MultiClass.__init__(self)
+
         self.eta = eta
         self.epochs = epochs
         self.l2 = l2

diff --git a/mlxtend/classifier/tests/test_adaline.py b/mlxtend/classifier/tests/test_adaline.py
@@ -79,6 +79,33 @@ def test_gradient_descent():
     assert((y1 == ada.predict(X_std)).all())
 
 
+def test_print_progress_1():
+    ada = Adaline(epochs=30,
+                  eta=0.01,
+                  minibatches=1,
+                  print_progress=1,
+                  random_seed=1)
+    ada.fit(X_std, y1)
+
+
+def test_print_progress_2():
+    ada = Adaline(epochs=30,
+                  eta=0.01,
+                  minibatches=1,
+                  print_progress=2,
+                  random_seed=1)
+    ada.fit(X_std, y1)
+
+
+def test_print_progress_3():
+    ada = Adaline(epochs=30,
+                  eta=0.01,
+                  minibatches=1,
+                  print_progress=3,
+                  random_seed=1)
+    ada.fit(X_std, y1)
+
+
 def test_score_function():
     ada = Adaline(epochs=30,
                   eta=0.01,

diff --git a/mlxtend/classifier/tests/test_logistic_regression.py b/mlxtend/classifier/tests/test_logistic_regression.py
@@ -67,6 +67,33 @@ def test_logistic_regression_gd():
     assert acc == 1.0, "Acc: %s" % acc
 
 
+def test_print_progress_1():
+    lr = LogisticRegression(epochs=100,
+                            eta=0.01,
+                            minibatches=1,
+                            print_progress=1,
+                            random_seed=1)
+    lr.fit(X, y)
+
+
+def test_print_progress_2():
+    lr = LogisticRegression(epochs=100,
+                            eta=0.01,
+                            minibatches=1,
+                            print_progress=2,
+                            random_seed=1)
+    lr.fit(X, y)
+
+
+def test_print_progress_3():
+    lr = LogisticRegression(epochs=100,
+                            eta=0.01,
+                            minibatches=1,
+                            print_progress=3,
+                            random_seed=1)
+    lr.fit(X, y)
+
+
 def test_score_function():
     lr = LogisticRegression(epochs=100,
                             eta=0.01,

diff --git a/mlxtend/classifier/tests/test_multilayerperceptron.py b/mlxtend/classifier/tests/test_multilayerperceptron.py
@@ -34,6 +34,36 @@ def test_multiclass_gd_acc():
     assert (y == mlp.predict(X)).all()
 
 
+def test_progress_1():
+    mlp = MLP(epochs=1,
+              eta=0.05,
+              hidden_layers=[10],
+              minibatches=1,
+              print_progress=1,
+              random_seed=1)
+    mlp.fit(X, y)
+
+
+def test_progress_2():
+    mlp = MLP(epochs=1,
+              eta=0.05,
+              hidden_layers=[10],
+              minibatches=1,
+              print_progress=2,
+              random_seed=1)
+    mlp.fit(X, y)
+
+
+def test_progress_3():
+    mlp = MLP(epochs=1,
+              eta=0.05,
+              hidden_layers=[10],
+              minibatches=1,
+              print_progress=3,
+              random_seed=1)
+    mlp.fit(X, y)
+
+
 def test_predict_proba():
     mlp = MLP(epochs=20,
               eta=0.05,

diff --git a/mlxtend/classifier/tests/test_perceptron.py b/mlxtend/classifier/tests/test_perceptron.py
@@ -59,6 +59,21 @@ def test_standardized_iris_data():
     assert (y0 == ppn.predict(X_std)).all(), ppn.predict(X_std)
 
 
+def test_progress_1():
+    ppn = Perceptron(epochs=15, eta=0.01, random_seed=1, print_progress=1)
+    ppn = ppn.fit(X_std, y0)
+
+
+def test_progress_2():
+    ppn = Perceptron(epochs=15, eta=0.01, random_seed=1, print_progress=2)
+    ppn = ppn.fit(X_std, y0)
+
+
+def test_progress_3():
+    ppn = Perceptron(epochs=15, eta=0.01, random_seed=1, print_progress=3)
+    ppn = ppn.fit(X_std, y0)
+
+
 def test_score_function():
     ppn = Perceptron(epochs=15, eta=0.01, random_seed=1)
     ppn = ppn.fit(X_std, y0)

diff --git a/mlxtend/classifier/tests/test_softmax_regression.py b/mlxtend/classifier/tests/test_softmax_regression.py
@@ -80,6 +80,36 @@ def test_binary_logistic_regression_sgd():
     assert (y_bin == lr.predict(X_bin)).all()
 
 
+def test_progress_1():
+    lr = SoftmaxRegression(epochs=1,
+                           eta=0.005,
+                           minibatches=1,
+                           print_progress=1,
+                           random_seed=1)
+
+    lr.fit(X_bin, y_bin)  # 0, 1 class
+
+
+def test_progress_2():
+    lr = SoftmaxRegression(epochs=1,
+                           eta=0.005,
+                           minibatches=1,
+                           print_progress=2,
+                           random_seed=1)
+
+    lr.fit(X_bin, y_bin)  # 0, 1 class
+
+
+def test_progress_3():
+    lr = SoftmaxRegression(epochs=1,
+                           eta=0.005,
+                           minibatches=1,
+                           print_progress=3,
+                           random_seed=1)
+
+    lr.fit(X_bin, y_bin)  # 0, 1 class
+
+
 def test_binary_l2_regularization_gd():
     t = np.array([[-0.17, 0.17],
                   [-2.26, 2.26]])

diff --git a/mlxtend/cluster/kmeans.py b/mlxtend/cluster/kmeans.py
@@ -58,6 +58,9 @@ def __init__(self, k, max_iter=10,
                  convergence_tolerance=1e-05,
                  random_seed=None, print_progress=0):
 
+        _BaseModel.__init__(self)
+        _Cluster.__init__(self)
+        _IterativeModel.__init__(self)
         self.k = k
         self.max_iter = max_iter
         self.convergence_tolerance = convergence_tolerance
@@ -71,7 +74,6 @@ def _fit(self, X, init_params=True):
         Called in self.fit
 
         """
-
         n_samples = X.shape[0]
 
         if init_params:

diff --git a/mlxtend/cluster/tests/test_kmeans.py b/mlxtend/cluster/tests/test_kmeans.py
@@ -35,6 +35,30 @@ def test_three_blobs_multi():
     assert (y_pred == y).all()
 
 
+def test_print_progress_1():
+    km = Kmeans(k=3,
+                max_iter=50,
+                random_seed=1,
+                print_progress=1)
+    km.fit(X)
+
+
+def test_print_progress_2():
+    km = Kmeans(k=3,
+                max_iter=50,
+                random_seed=1,
+                print_progress=2)
+    km.fit(X)
+
+
+def test_print_progress_3():
+    km = Kmeans(k=3,
+                max_iter=50,
+                random_seed=1,
+                print_progress=3)
+    km.fit(X)
+
+
 def test_three_blobs_1sample():
     km = Kmeans(k=3,
                 max_iter=50,

diff --git a/mlxtend/regressor/linear_regression.py b/mlxtend/regressor/linear_regression.py
@@ -55,6 +55,9 @@ def __init__(self, eta=0.01, epochs=50,
                  minibatches=None, random_seed=None,
                  print_progress=0):
 
+        _BaseModel.__init__(self)
+        _IterativeModel.__init__(self)
+        _Regressor.__init__(self)
         self.eta = eta
         self.epochs = epochs
         self.minibatches = minibatches