Merge 2b696b8 into 9a8b3c9

.travis.yml

@@ -36,6 +36,7 @@ before_script:
 
 script:
   - coverage${PYVERSION} run --source ffx tests/x_square_test.py
+  - python tests/test_sklearn_api.py
 
 after_success:
   - coveralls

ffx/api.py

@@ -1,27 +1,40 @@
 from sklearn.base import BaseEstimator, RegressorMixin
+from sklearn.utils import check_array, check_X_y
+from sklearn.utils.validation import check_is_fitted
 from . import core
 
-"""api.py defines user interfaces to FFX. run() runs the complete method.
-FFXRegressor is a Scikit-learn style regressor."""
+""" api.py defines user interfaces to FFX. run() runs the complete method.
+FFXRegressor is a Scikit-learn style regressor.
+"""
+
 
 def run(train_X, train_y, test_X, test_y, varnames=None, verbose=False):
-    return core.MultiFFXModelFactory().build(train_X, train_y, test_X, test_y, varnames, verbose)
+    return core.MultiFFXModelFactory().build(train_X, train_y, test_X, test_y,
+                                             varnames, verbose)
+
 
 class FFXRegressor(BaseEstimator, RegressorMixin):
     """This class provides a Scikit-learn style estimator."""
+
     def __init__(self):
         pass
+
     def fit(self, X, y):
+        X, y = check_X_y(X, y, y_numeric=True, multi_output=False)
         # if X is a Pandas DataFrame, we don't have to pass in varnames.
-        # otherwise we make up placeholders.        
+        # otherwise we make up placeholders.
         if hasattr(X, 'columns'):
             varnames = None
         else:
             varnames = ["X%d" % i for i in range(len(X))]
         self._models = run(X, y, X, y, varnames=varnames)
         self._model = self._models[-1]
+        return self
+
     def predict(self, X):
+        check_is_fitted(self, "_model")
+        X = check_array(X, accept_sparse=False)
         return self._model.simulate(X)
+
     def complexity(self):
         return self._model.complexity()
-

ffx/core.py

@@ -702,8 +702,10 @@ def build(self, X, y, ss, varnames=None, verbose=False):
                         minx, maxx = min(X[:, var_i]), max(X[:, var_i])
                         rangex = maxx - minx
                         stepx = 0.8 * rangex / float(ss.num_thrs_per_var + 1)
-                        thrs = numpy.arange(
-                            minx + 0.2 * rangex, maxx - 0.2 * rangex + 0.1 * rangex, stepx)
+                        thrs = [maxx]
+                        if rangex > 0:
+                            thrs = numpy.arange(
+                                minx + 0.2 * rangex, maxx - 0.2 * rangex + 0.1 * rangex, stepx)
                         for threshold_op in ss.thresholdOps():
                             for thr in thrs:
                                 nonsimple_base = OperatorBase(

tests/test_sklearn_api.py

@@ -2,6 +2,7 @@
 
 import numpy as np
 import ffx
+from sklearn.utils.estimator_checks import check_estimator
 
 # This creates a dataset of 2 predictors
 X = np.random.random((20, 2))
@@ -16,3 +17,6 @@
 print("Score:", FFX.score(test_X, test_y))
 print("Complexity:", FFX.complexity())
 print("Model:", FFX._model)
+
+# This tests the regressor using sklearn's test facilities
+check_estimator(ffx.FFXRegressor)