* #8477
check length of feature_names in export_graphviz
- raise ValueError if len(feature_names) > tree.n_features
- add unit test for len(feature_names) > tree.n_features
- change the comment of existing unit test for len(feature_names) <
tree.n_features
* fix error and warning
- include length of feature_names and number of features in tree in the
error and warning message.
- raise error for too few feature_names
- for too much feature_names, will use the first n_features. raise an
warning for users
- use assert_raise_message and assert_warns_message in test to check
message.
* move the error and warning from node_to_str to export_graphviz so it will fail early for wrong length of feature_names
* raise error if length of feature_names does not match number of features in the decision tree
* fix pep8
* remove unused assert_warns_message import in test_export.py
* add bug fix in doc/whats_new.rst
* fix the english in doc/whats_new.rst
* fix the format and english in sklearn/tree/export.py
* fix contributor format in doc/whats_new.rst
* fix english, use bracket and avoid \ in error message
* fix pep8
* fix pep8
Replaced all BSD style licenses with "BSD 3 clause"
Not checked yet!
Removed duplicate "3 clause, 3 clause"
Removed trailing period if exists
Fixed some missed licences, still about 50 to do, but those can be automated
Think I got the last of them.
Apparently me and sed have different ideas of regex.
Found a few more