Merge branch 'main' into python39

docs/source/conf.py

@@ -152,7 +152,7 @@
 # documentation.
 html_theme_options = {
     "github_url": "https://github.com/alteryx/featuretools",
-    "twitter_url": "https://twitter.com/featuretools_py"
+    "twitter_url": "https://twitter.com/AlteryxOSS",
 }
 
 # Add any paths that contain custom themes here, relative to this directory.

docs/source/release_notes.rst

@@ -6,15 +6,18 @@ Release Notes
     * Enhancements
        * Add Python 3.9 support without Koalas testing (:pr:`1318`)
     * Fixes
+        * Calculate direct features uses default value if parent missing (:pr:`1312`)
     * Changes
     * Documentation Changes
+        * Update Twitter link to documentation toolbar (:pr:`1322`)
     * Testing Changes
         * Unpin python-graphviz package on Windows (:pr:`1296`)
         * Reorganize and clean up tests (:pr:`1294`, :pr:`1303`, :pr:`1306`)
         * Trigger tests on pull request events (:pr:`1304`, :pr:`1315`)
+        * Remove unnecessary test skips on Windows (:pr:`1320`)
 
     Thanks to the following people for contributing to this release:
-    :user:`gsheni`, :user:`jeff-hernandez`, :user:`rwedge`, :user:`thehomebrewnerd`
+    :user:`gsheni`, :user:`jeff-hernandez`, :user:`rwedge`, :user:`thehomebrewnerd`, :user:`seriallazer`
 
 **v0.23.0 Dec 31, 2020**
     * Fixes

docs/source/templates/layout.html

@@ -39,7 +39,7 @@
       <a href="https://github.com/alteryx/featuretools#readme" target="_blank">
         <img class="footer-image-github" src="{{ pathto('_static/images/github.svg', 1) }}" alt="GitHub">
       </a>
-      <a href="https://twitter.com/featuretools_py?lang=en" target="_blank">
+      <a href="https://twitter.com/AlteryxOSS" target="_blank">
         <img class="footer-image-twitter" src="{{ pathto('_static/images/twitter.svg', 1) }}" alt="Twitter">
       </a>
     </div>

featuretools/computational_backends/feature_set_calculator.py

@@ -538,7 +538,12 @@ def _calculate_direct_features(self, features, child_df, df_trie, progress_callb
         # new column names (in the child entity) for the merge
         col_map = {relationship.parent_variable.id: merge_var}
         index_as_feature = None
+
+        fillna_dict = {}
         for f in features:
+            feature_defaults = {name: f.default_value
+                                for name in f.get_feature_names() if not pd.isna(f.default_value)}
+            fillna_dict.update(feature_defaults)
             if f.base_features[0].get_name() == relationship.parent_variable.id:
                 index_as_feature = f
             base_names = f.base_features[0].get_feature_names()
@@ -565,7 +570,7 @@ def _calculate_direct_features(self, features, child_df, df_trie, progress_callb
 
         progress_callback(len(features) / float(self.num_features))
 
-        return new_df
+        return new_df.fillna(fillna_dict)
 
     def _calculate_agg_features(self, features, frame, df_trie, progress_callback):
         test_feature = features[0]

featuretools/tests/computational_backend/test_calculate_feature_matrix.py

@@ -867,10 +867,12 @@ def test_empty_path_approximate_full(pd_es):
                                               approximate=Timedelta(10, 's'),
                                               cutoff_time=cutoff_time)
     vals1 = feature_matrix[dfeat.get_name()].tolist()
-    assert np.isnan(vals1[0])
-    assert np.isnan(vals1[1])
+
+    assert (vals1[0] == 0)
+    assert (vals1[1] == 0)
     assert feature_matrix[agg_feat.get_name()].tolist() == [5, 1]
 
+
 # todo: do we need to test this situation?
 # def test_empty_path_approximate_partial(pd_es):
 #     pd_es = copy.deepcopy(pd_es)
@@ -1828,6 +1830,21 @@ def test_calc_feature_matrix_with_cutoff_df_and_instance_ids(es):
     assert (feature_matrix[property_feature.get_name()] == labels).values.all()
 
 
+def test_calculate_feature_matrix_returns_default_values(default_value_es):
+    sum_features = ft.Feature(default_value_es["transactions"]["value"],
+                              parent_entity=default_value_es["sessions"], primitive=Sum)
+    sessions_sum = ft.Feature(sum_features,
+                              entity=default_value_es["transactions"])
+
+    feature_matrix = ft.calculate_feature_matrix(features=[sessions_sum],
+                                                 entityset=default_value_es)
+
+    feature_matrix = to_pandas(feature_matrix, index='id', sort_index=True)
+    expected_values = [2.0, 2.0, 1.0, 0.0]
+
+    assert (feature_matrix[sessions_sum.get_name()] == expected_values).values.all()
+
+
 def test_entities_relationships(entities, relationships):
     fm_1, features = ft.dfs(entities=entities,
                             relationships=relationships,

featuretools/tests/computational_backend/test_feature_set_calculator.py

@@ -1,4 +1,3 @@
-import sys
 from datetime import datetime
 
 import numpy as np
@@ -439,8 +438,6 @@ def dd_df(pd_df):
 @pytest.fixture
 def ks_df(pd_df):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_df)
 
 
@@ -749,8 +746,6 @@ def dd_parent_child(pd_parent_child):
 @pytest.fixture
 def ks_parent_child(pd_parent_child):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     parent_df, child_df = pd_parent_child
     parent_df = ks.from_pandas(parent_df)
     child_df = ks.from_pandas(child_df)

featuretools/tests/conftest.py

@@ -1,5 +1,4 @@
 import copy
-import sys
 
 import composeml as cp
 import dask.dataframe as dd
@@ -60,8 +59,6 @@ def dask_es(make_es):
 @pytest.fixture
 def ks_es(make_es):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     ks_es = copy.deepcopy(make_es)
     for entity in ks_es.entities:
         cleaned_df = pd_to_ks_clean(entity.df).reset_index(drop=True)
@@ -143,8 +140,6 @@ def dask_diamond_es(pd_diamond_es):
 @pytest.fixture
 def ks_diamond_es(pd_diamond_es):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     entities = {}
     for entity in pd_diamond_es.entities:
         entities[entity.id] = (ks.from_pandas(pd_to_ks_clean(entity.df)), entity.index, None, entity.variable_types)
@@ -157,6 +152,64 @@ def ks_diamond_es(pd_diamond_es):
     return ft.EntitySet(id=pd_diamond_es.id, entities=entities, relationships=relationships)
 
 
+@pytest.fixture(params=['pd_default_value_es', 'dask_default_value_es', 'ks_default_value_es'])
+def default_value_es(request):
+    return request.getfixturevalue(request.param)
+
+
+@pytest.fixture
+def pd_default_value_es():
+    transactions = pd.DataFrame({
+        "id": [1, 2, 3, 4],
+        "session_id": ["a", "a", "b", "c"],
+        "value": [1, 1, 1, 1]
+    })
+
+    sessions = pd.DataFrame({
+        "id": ["a", "b"]
+    })
+
+    es = ft.EntitySet()
+    es.entity_from_dataframe(entity_id="transactions",
+                             dataframe=transactions,
+                             index="id")
+    es.entity_from_dataframe(entity_id="sessions",
+                             dataframe=sessions,
+                             index="id")
+
+    es.add_relationship(ft.Relationship(es["sessions"]["id"], es["transactions"]["session_id"]))
+    return es
+
+
+@pytest.fixture
+def dask_default_value_es(pd_default_value_es):
+    entities = {}
+    for entity in pd_default_value_es.entities:
+        entities[entity.id] = (dd.from_pandas(entity.df, npartitions=4), entity.index, None, entity.variable_types)
+
+    relationships = [(rel.parent_entity.id,
+                      rel.parent_variable.name,
+                      rel.child_entity.id,
+                      rel.child_variable.name) for rel in pd_default_value_es.relationships]
+
+    return ft.EntitySet(id=pd_default_value_es.id, entities=entities, relationships=relationships)
+
+
+@pytest.fixture
+def ks_default_value_es(pd_default_value_es):
+    ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
+    entities = {}
+    for entity in pd_default_value_es.entities:
+        entities[entity.id] = (ks.from_pandas(pd_to_ks_clean(entity.df)), entity.index, None, entity.variable_types)
+
+    relationships = [(rel.parent_entity.id,
+                      rel.parent_variable.name,
+                      rel.child_entity.id,
+                      rel.child_variable.name) for rel in pd_default_value_es.relationships]
+
+    return ft.EntitySet(id=pd_default_value_es.id, entities=entities, relationships=relationships)
+
+
 @pytest.fixture(params=['pd_home_games_es', 'dask_home_games_es', 'ks_home_games_es'])
 def home_games_es(request):
     return request.getfixturevalue(request.param)
@@ -198,8 +251,6 @@ def dask_home_games_es(pd_home_games_es):
 @pytest.fixture
 def ks_home_games_es(pd_home_games_es):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     entities = {}
     for entity in pd_home_games_es.entities:
         entities[entity.id] = (ks.from_pandas(pd_to_ks_clean(entity.df)), entity.index, None, entity.variable_types)
@@ -235,8 +286,6 @@ def dd_mock_customer(pd_mock_customer):
 @pytest.fixture
 def ks_mock_customer(pd_mock_customer):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     ks_mock_customer = copy.deepcopy(pd_mock_customer)
     for entity in ks_mock_customer.entities:
         cleaned_df = pd_to_ks_clean(entity.df).reset_index(drop=True)
@@ -320,8 +369,6 @@ def dask_entities():
 @pytest.fixture
 def koalas_entities():
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     cards_df = ks.DataFrame({"id": [1, 2, 3, 4, 5]})
     transactions_df = ks.DataFrame({"id": [1, 2, 3, 4, 5, 6],
                                     "card_id": [1, 2, 1, 3, 4, 5],

featuretools/tests/entityset_tests/test_es.py

@@ -1,5 +1,4 @@
 import copy
-import sys
 from datetime import datetime
 
 import dask.dataframe as dd
@@ -219,8 +218,6 @@ def dd_df(pd_df):
 @pytest.fixture
 def ks_df(pd_df):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_df)
 
 
@@ -286,8 +283,6 @@ def dd_df2(pd_df2):
 @pytest.fixture
 def ks_df2(pd_df2):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_df2)
 
 
@@ -322,8 +317,6 @@ def dd_df3(pd_df3):
 @pytest.fixture
 def ks_df3(pd_df3):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_df3)
 
 
@@ -384,8 +377,6 @@ def dd_df4(pd_df4):
 @pytest.fixture
 def ks_df4(pd_df4):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_to_ks_clean(pd_df4))
 
 
@@ -481,8 +472,6 @@ def dd_datetime1(pd_datetime1):
 @pytest.fixture
 def ks_datetime1(pd_datetime1):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_datetime1)
 
 
@@ -527,8 +516,6 @@ def dd_datetime2(pd_datetime2):
 @pytest.fixture
 def ks_datetime2(pd_datetime2):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_datetime2)
 
 
@@ -787,8 +774,6 @@ def dd_transactions_df(pd_transactions_df):
 @pytest.fixture
 def ks_transactions_df(pd_transactions_df):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_transactions_df)
 
 
@@ -1227,8 +1212,6 @@ def dd_datetime3(pd_datetime3):
 @pytest.fixture
 def ks_datetime3(pd_datetime3):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_datetime3)
 
 
@@ -1277,8 +1260,6 @@ def dd_index_df(pd_index_df):
 @pytest.fixture
 def ks_index_df(pd_index_df):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_index_df)
 
 

featuretools/tests/entityset_tests/test_es_metadata.py

@@ -1,5 +1,3 @@
-import sys
-
 import pandas as pd
 import pytest
 from dask import dataframe as dd
@@ -150,8 +148,6 @@ def dd_employee_df(pd_employee_df):
 @pytest.fixture
 def ks_employee_df(pd_employee_df):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     return ks.from_pandas(pd_employee_df)
 
 

featuretools/tests/entityset_tests/test_plotting.py

@@ -1,6 +1,5 @@
 import os
 import re
-import sys
 
 import graphviz
 import pandas as pd
@@ -30,8 +29,6 @@ def dd_simple():
 @pytest.fixture
 def ks_simple():
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     es = ft.EntitySet("test")
     df = ks.DataFrame({'foo': [1]})
     es.entity_from_dataframe('test', df)

featuretools/tests/synthesis/test_deep_feature_synthesis.py

@@ -1,5 +1,4 @@
 import copy
-import sys
 
 import dask.dataframe as dd
 import pandas as pd
@@ -85,8 +84,6 @@ def dask_transform_es(pd_transform_es):
 @pytest.fixture
 def koalas_transform_es(pd_transform_es):
     ks = pytest.importorskip('databricks.koalas', reason="Koalas not installed, skipping")
-    if sys.platform.startswith('win'):
-        pytest.skip('skipping Koalas tests for Windows')
     es = ft.EntitySet(id=pd_transform_es.id)
     for entity in pd_transform_es.entities:
         es.entity_from_dataframe(entity_id=entity.id,