Updated feature_extraction doc

rth · rth · commit 02bff9eea43f · 2017-06-20T15:46:52.000+03:00
diff --git a/doc/modules/feature_extraction.rst b/doc/modules/feature_extraction.rst
@@ -125,11 +125,10 @@ Since the hash function might cause collisions between (unrelated) features,
 a signed hash function is used and the sign of the hash value
 determines the sign of the value stored in the output matrix for a feature.
 This way, collisions are likely to cancel out rather than accumulate error,
-and the expected mean of any output feature's value is zero.
-
-If ``non_negative=True`` is passed to the constructor, the absolute
-value is taken.  This undoes some of the collision handling, but allows
-the output to be passed to estimators like
+and the expected mean of any output feature's value is zero. This mechanism
+is enabled by default with ``alternate_sign=True`` and is particularly useful
+for small hash table sizes (``n_features < 10000``). For large hash table
+sizes, it can be disabled to allow the output to be passed to estimators like
 :class:`sklearn.naive_bayes.MultinomialNB` or
 :class:`sklearn.feature_selection.chi2`
 feature selectors that expect non-negative inputs.
