Merge 57eb82c into 05bc785

CHANGELOG.md

@@ -2,7 +2,8 @@
 
 #### Forthcoming 0.9.0
  - new prediction function in `CoxPHFitter`, `predict_log_hazard_relative_to_mean`, that mimics what R's `predict.coxph` does.
- - changing the default `predict` method in lifelines to return _not the median_, but another value dependent on the fitter that is calling it. This is because too often the `predict_median` function was returning inf values which would significantly damage measure of concordence index. 
+ - removing the `predict` method in CoxPHFitter and AalenAdditiveFitter. This is because the choice of `predict_median` as a default was causing too much confusion, and no other natual choice as a default was available. All other `predict_` methods remain. 
+ - Default predict method in `k_fold_cross_validation` is now `predict_expectation`
 
 #### 0.8.1
  - supports matplotlib 1.5.

lifelines/fitters/aalen_additive_fitter.py

@@ -414,17 +414,6 @@ def predict_expectation(self, X):
         t = self.cumulative_hazards_.index
         return pd.DataFrame(trapz(self.predict_survival_function(X)[index].values.T, t), index=index)
 
-    def predict(self, X):
-        """
-        X: a (n,d) covariate numpy array or DataFrame. If a DataFrame, columns
-            can be in any order. If a numpy array, columns must be in the
-            same order as the training data.
-
-        Returns the median lifetimes for the individuals. Alias for predict_median
-        """
-        return self.predict_median(X)
-
-
     def plot(self, ix=None, iloc=None, columns=[], legend=True, **kwargs):
         """"
         A wrapper around plotting. Matplotlib plot arguments can be passed in, plus:
@@ -476,4 +465,4 @@ def shaded_plot(ax, x, y, y_upper, y_lower, **kwargs):
         if legend:
             ax.legend()
 
-        return ax
+        return ax

lifelines/fitters/coxph_fitter.py

@@ -497,18 +497,6 @@ def predict_expectation(self, X):
         v = self.predict_survival_function(X)[index]
         return pd.DataFrame(trapz(v.values.T, v.index), index=index)
 
-    def predict(self, X):
-        """
-        X: a (n,d) covariate numpy array or DataFrame. If a DataFrame, columns
-            can be in any order. If a numpy array, columns must be in the
-            same order as the training data.        
-
-        Returns the median predicted lifetime. This is different from R's predict.coxph, which returns
-        the  the linear predictor for the log-hazard relative to a mean survival estimate (this is 
-        available in predict_log_hazard_relative_to_mean)
-        """
-        return self.predict_median(X)
-
     def _compute_baseline_hazard(self):
         # http://courses.nus.edu.sg/course/stacar/internet/st3242/handouts/notes3.pdf
         ind_hazards = self.predict_partial_hazard(self.data).values

lifelines/utils/__init__.py

@@ -416,7 +416,7 @@ def AandS_approximation(p):
 
 def k_fold_cross_validation(fitters, df, duration_col, event_col=None,
                             k=5, evaluation_measure=concordance_index,
-                            predictor="predict_median", predictor_kwargs={}):
+                            predictor="predict_expectation", predictor_kwargs={}):
     """
     Perform cross validation on a dataset. If multiple models are provided,
     all models will train on each of the k subsets.
@@ -440,7 +440,7 @@ def k_fold_cross_validation(fitters, df, duration_col, event_col=None,
                         between two series of event times
     predictor: a string that matches a prediction method on the fitter instances.
                For example, "predict_expectation" or "predict_percentile".
-               Default is "predict_median"
+               Default is "predict_expectation"
                The interface for the method is:
                    predict(self, data, **optional_kwargs)
     predictor_kwargs: keyword args to pass into predictor-method.

tests/test_estimation.py

@@ -565,7 +565,7 @@ def test_predict_methods_in_regression_return_same_types(self):
         aaf.fit(X, duration_col='T', event_col='E')
         cph.fit(X, duration_col='T', event_col='E')
 
-        for fit_method in ['predict_percentile', 'predict_median', 'predict_expectation', 'predict_survival_function', 'predict', 'predict_cumulative_hazard']:
+        for fit_method in ['predict_percentile', 'predict_median', 'predict_expectation', 'predict_survival_function', 'predict_cumulative_hazard']:
             assert isinstance(getattr(aaf, fit_method)(X), type(getattr(cph, fit_method)(X)))
 
     def test_duration_vector_can_be_normalized(self):
@@ -588,13 +588,11 @@ def test_prediction_methods_respect_index(self, data_pred2):
         cph.fit(data_pred2, duration_col='t', event_col='E')
         npt.assert_array_equal(cph.predict_partial_hazard(x).index, expected_index)
         npt.assert_array_equal(cph.predict_percentile(x).index, expected_index)
-        npt.assert_array_equal(cph.predict(x).index, expected_index)
         npt.assert_array_equal(cph.predict_expectation(x).index, expected_index)
 
         aaf = AalenAdditiveFitter()
         aaf.fit(data_pred2, duration_col='t', event_col='E')
         npt.assert_array_equal(aaf.predict_percentile(x).index, expected_index)
-        npt.assert_array_equal(aaf.predict(x).index, expected_index)
         npt.assert_array_equal(aaf.predict_expectation(x).index, expected_index)
 
 
@@ -1145,7 +1143,7 @@ def test_crossval_for_aalen_add(self, data_pred2, data_pred1):
 
             expected = 0.90
             msg = "Expected min-mean c-index {:.2f} < {:.2f}"
-            assert np.mean(mean_scores) > expected, msg.format(expected, scores.mean())
+            assert np.mean(mean_scores) > expected, msg.format(expected, np.mean(scores))
 
     def test_predict_cumulative_hazard_inputs(self, data_pred1):
         aaf = AalenAdditiveFitter()