MAINT remove python2 support (#623)

* MAINT remove python2 support

* MAINT reduce the amount of warnings

* MAINT PEP8

* MAINT improve style

Author: mfeurer

.travis.yml

@@ -15,7 +15,6 @@ env:
   - TEST_DIR=/tmp/test_dir/
   - MODULE=openml
   matrix:
-  - DISTRIB="conda" PYTHON_VERSION="2.7" SKLEARN_VERSION="0.20.0"
   - DISTRIB="conda" PYTHON_VERSION="3.5" SKLEARN_VERSION="0.20.0"
   - DISTRIB="conda" PYTHON_VERSION="3.6" SKLEARN_VERSION="0.20.0"
   - DISTRIB="conda" PYTHON_VERSION="3.7" SKLEARN_VERSION="0.20.0" RUN_FLAKE8="true" SKIP_TESTS="true"

ci_scripts/flake8_diff.sh

@@ -38,6 +38,13 @@ echo "Remotes:"
 echo '--------------------------------------------------------------------------------'
 git remote --verbose
 
+echo "Travis variables:"
+echo '--------------------------------------------------------------------------------'
+echo "On travis: $TRAVIS"
+echo "Current branch: $TRAVIS_BRANCH"
+echo "Is a pull request test: $TRAVIS_PULL_REQUEST"
+echo "Repository: $TRAVIS_REPO_SLUG"
+
 # Travis does the git clone with a limited depth (50 at the time of
 # writing). This may not be enough to find the common ancestor with
 # $REMOTE/develop so we unshallow the git checkout
@@ -48,6 +55,14 @@ if [[ -a .git/shallow ]]; then
 fi
 
 if [[ "$TRAVIS" == "true" ]]; then
+    if [[ "$TRAVIS_BRANCH" == "master" ]]
+    then
+        # We do not test PEP8 on the master branch (or for the PR test into
+        # master) as this results in failures which are only shown for the
+        # pull request to finish a release (development to master) and are
+        # therefore a pain to fix
+        exit 0
+    fi
     if [[ "$TRAVIS_PULL_REQUEST" == "false" ]]
     then
         # In main repo, using TRAVIS_COMMIT_RANGE to test the commits
@@ -116,7 +131,7 @@ echo -e '\nRunning flake8 on the diff in the range' "$COMMIT_RANGE" \
 echo '--------------------------------------------------------------------------------'
 # We need the following command to exit with 0 hence the echo in case
 # there is no match
-MODIFIED_FILES="$(git diff --name-only $COMMIT_RANGE || echo "no_match")"
+MODIFIED_FILES="$(git diff --no-ext-diff --name-only $COMMIT_RANGE || echo "no_match")"
 
 check_files() {
     files="$1"
@@ -125,7 +140,7 @@ check_files() {
     if [ -n "$files" ]; then
         # Conservative approach: diff without context (--unified=0) so that code
         # that was not changed does not create failures
-        git diff --unified=0 $COMMIT_RANGE -- $files | flake8 --ignore E402 --diff --show-source $options
+        git diff --no-ext-diff --unified=0 $COMMIT_RANGE -- $files | flake8 --ignore E402 --diff --show-source $options
     fi
 }
 
@@ -137,4 +152,4 @@ else
     check_files "$(echo "$MODIFIED_FILES" | grep ^examples)" \
         --config ./examples/.flake8
 fi
-echo -e "No problem detected by flake8\n"
+echo -e "No problem detected by flake8\n"

doc/index.rst

@@ -22,11 +22,11 @@ Example
 
     import openml
     from sklearn import preprocessing, tree, pipeline
-   
+
     # Set the OpenML API Key which is required to upload your runs.
     # You can get your own API by signing up to OpenML.org.
     openml.config.apikey = 'ABC'
-    
+
     # Define a scikit-learn classifier or pipeline
     clf = pipeline.Pipeline(
         steps=[
@@ -38,7 +38,7 @@ Example
     # cross-validation.
     task = openml.tasks.get_task(31)
     # Run the scikit-learn model on the task (requires an API key).
-    run = openml.runs.run_model_on_task(task, clf)
+    run = openml.runs.run_model_on_task(clf, task)
     # Publish the experiment on OpenML (optional, requires an API key).
     run.publish()
     print('View the run online: %s/run/%d' % (openml.config.server, run.run_id))

doc/progress.rst

@@ -9,6 +9,12 @@ Progress
 Changelog
 =========
 
+0.9.0
+~~~~~
+
+* ADD #560: OpenML-Python can now handle regression tasks as well.
+* MAINT #184: Dropping Python2 support.
+
 0.8.0
 ~~~~~
 

openml/config.py

@@ -4,9 +4,9 @@
 import logging
 import os
 
-from six import StringIO
-from six.moves import configparser
-from six.moves.urllib_parse import urlparse
+from io import StringIO
+import configparser
+from urllib.parse import urlparse
 
 
 logger = logging.getLogger(__name__)
@@ -91,7 +91,7 @@ def _parse_config():
             for line in fh:
                 config_file_.write(line)
         config_file_.seek(0)
-        config.readfp(config_file_)
+        config.read_file(config_file_)
     except OSError as e:
         logging.info("Error opening file %s: %s", config_file, e.message)
     return config

openml/datasets/data_feature.py

@@ -1,5 +1,3 @@
-import six
-
 class OpenMLDataFeature(object):
     """Data Feature (a.k.a. Attribute) object.
 
@@ -30,11 +28,7 @@ def __init__(self, index, name, data_type, nominal_values,
             raise ValueError('number_missing_values is of wrong datatype')
 
         self.index = index
-        # In case of python version lower than 3, change the default ASCII encoder.
-        if six.PY2:
-            self.name = str(name.encode('utf8'))
-        else:
-            self.name = str(name)
+        self.name = str(name)
         self.data_type = str(data_type)
         self.nominal_values = nominal_values
         self.number_missing_values = number_missing_values

openml/datasets/dataset.py

@@ -2,14 +2,13 @@
 import io
 import logging
 import os
+import pickle
 from collections import OrderedDict
 
 import arff
 import numpy as np
 import scipy.sparse
 import xmltodict
-import six
-from six.moves import cPickle as pickle
 from warnings import warn
 
 import openml._api_calls
@@ -122,7 +121,7 @@ def __init__(self, name, description, format=None,
         self.default_target_attribute = default_target_attribute
         self.row_id_attribute = row_id_attribute
         self.ignore_attributes = None
-        if isinstance(ignore_attribute, six.string_types):
+        if isinstance(ignore_attribute, str):
             self.ignore_attributes = [ignore_attribute]
         elif isinstance(ignore_attribute, list):
             self.ignore_attributes = ignore_attribute
@@ -159,10 +158,7 @@ def __init__(self, name, description, format=None,
 
         if data_file is not None:
             if self._data_features_supported():
-                if six.PY2:
-                    self.data_pickle_file = data_file.replace('.arff', '.pkl.py2')
-                else:
-                    self.data_pickle_file = data_file.replace('.arff', '.pkl.py3')
+                self.data_pickle_file = data_file.replace('.arff', '.pkl.py3')
 
                 if os.path.exists(self.data_pickle_file):
                     logger.debug("Data pickle file already exists.")
@@ -327,7 +323,7 @@ def get_data(self, target=None,
             if not self.row_id_attribute:
                 pass
             else:
-                if isinstance(self.row_id_attribute, six.string_types):
+                if isinstance(self.row_id_attribute, str):
                     to_exclude.append(self.row_id_attribute)
                 else:
                     to_exclude.extend(self.row_id_attribute)
@@ -336,7 +332,7 @@ def get_data(self, target=None,
             if not self.ignore_attributes:
                 pass
             else:
-                if isinstance(self.ignore_attributes, six.string_types):
+                if isinstance(self.ignore_attributes, str):
                     to_exclude.append(self.ignore_attributes)
                 else:
                     to_exclude.extend(self.ignore_attributes)
@@ -354,7 +350,7 @@ def get_data(self, target=None,
         if target is None:
             rval.append(data)
         else:
-            if isinstance(target, six.string_types):
+            if isinstance(target, str):
                 if ',' in target:
                     target = target.split(',')
                 else:
@@ -368,7 +364,7 @@ def get_data(self, target=None,
                 )
             target_categorical = [
                 cat for cat, column in
-                six.moves.zip(categorical, attribute_names)
+                zip(categorical, attribute_names)
                 if column in target
             ]
             target_dtype = int if target_categorical[0] else float
@@ -475,7 +471,7 @@ def get_features_by_type(self, data_type, exclude=None,
             if not isinstance(self.ignore_attributes, list):
                 raise TypeError("ignore_attributes should be a list")
         if self.row_id_attribute is not None:
-            if not isinstance(self.row_id_attribute, six.string_types):
+            if not isinstance(self.row_id_attribute, str):
                 raise TypeError("row id attribute should be a str")
         if exclude is not None:
             if not isinstance(exclude, list):

openml/datasets/functions.py

@@ -2,17 +2,20 @@
 import io
 import os
 import re
+import warnings
 
 import numpy as np
-import six
 import arff
 import pandas as pd
 
 import xmltodict
 from scipy.sparse import coo_matrix
-from oslo_concurrency import lockutils
+# Currently, importing oslo raises a lot of warning that it will stop working
+# under python3.8; remove this once they disappear
+with warnings.catch_warnings():
+    warnings.simplefilter("ignore")
+    from oslo_concurrency import lockutils
 from collections import OrderedDict
-from warnings import warn
 
 import openml.utils
 import openml._api_calls
@@ -348,7 +351,7 @@ def get_dataset(dataset_id):
         except OpenMLServerException as e:
             # if there was an exception, check if the user had access to the dataset
             if e.code == 112:
-                six.raise_from(PrivateDatasetError(e.message), None)
+                raise PrivateDatasetError(e.message) from None
             else:
                 raise e
         finally:

openml/flows/flow.py

@@ -1,6 +1,5 @@
 from collections import OrderedDict
 
-import six
 import xmltodict
 
 import openml._api_calls
@@ -192,14 +191,15 @@ def _to_dict(self):
                              meta_info['description'])
 
             for key_, value in param_dict.items():
-                if key_ is not None and not isinstance(key_, six.string_types):
+                if key_ is not None and not isinstance(key_, str):
                     raise ValueError('Parameter name %s cannot be serialized '
                                      'because it is of type %s. Only strings '
                                      'can be serialized.' % (key_, type(key_)))
-                if value is not None and not isinstance(value, six.string_types):
+                if value is not None and not isinstance(value, str):
                     raise ValueError('Parameter value %s cannot be serialized '
                                      'because it is of type %s. Only strings '
-                                     'can be serialized.' % (value, type(value)))
+                                     'can be serialized.'
+                                     % (value, type(value)))
 
             flow_parameters.append(param_dict)
 
@@ -215,7 +215,7 @@ def _to_dict(self):
             for key_ in component_dict:
                 # We only need to check if the key is a string, because the
                 # value is a flow. The flow itself is valid by recursion
-                if key_ is not None and not isinstance(key_, six.string_types):
+                if key_ is not None and not isinstance(key_, str):
                     raise ValueError('Parameter name %s cannot be serialized '
                                      'because it is of type %s. Only strings '
                                      'can be serialized.' % (key_, type(key_)))

openml/flows/functions.py

@@ -1,7 +1,6 @@
 import dateutil.parser
 
 import xmltodict
-import six
 
 import openml._api_calls
 from . import OpenMLFlow
@@ -119,9 +118,9 @@ def flow_exists(name, external_version):
     -----
     see http://www.openml.org/api_docs/#!/flow/get_flow_exists_name_version
     """
-    if not (isinstance(name, six.string_types) and len(name) > 0):
+    if not (isinstance(name, str) and len(name) > 0):
         raise ValueError('Argument \'name\' should be a non-empty string')
-    if not (isinstance(name, six.string_types) and len(external_version) > 0):
+    if not (isinstance(name, str) and len(external_version) > 0):
         raise ValueError('Argument \'version\' should be a non-empty string')
 
     xml_response = openml._api_calls._perform_api_call(