Scheduled monthly dependency update for July (#789)

* Update bumpversion from 0.5.3 to 0.6.0 * Update mkdocs from 1.0.4 to 1.1.2 * Update pymdown-extensions from 6.0 to 7.1 * Update mkdocs-material from 4.2.0 to 5.4.0 * Update black from 19.3b0 to 19.10b0 * Update sqlalchemy from 1.3.17 to 1.3.18 * Update boto3 from 1.13.21 to 1.14.14 * Update inflection from 0.4.0 to 0.5.0 * Update numpy from 1.18.4 to 1.19.0 * Update requests from 2.23.0 to 2.24.0 * Update scipy from 1.3.1 to 1.5.0 * Update scikit-learn from 0.20.3 to 0.23.1 * Update matplotlib from 3.2.1 to 3.2.2 * Update pandas from 1.0.4 to 1.0.5 * Update pytest-cov from 2.9.0 to 2.10.0 * Update hypothesis from 5.16.0 to 5.19.0 * Update flake8 from 3.7.7 to 3.8.3 * Update codecov from 2.0.17 to 2.1.7 * Update tox from 3.9.0 to 3.16.1 * New way of import joblib * LogisticRegression is now *only* in sklearn.linear_model * Updated ScaledLogisticRegression to the new default in scikitlearn (solver = lbfgs). Improved message in the transformer. Co-authored-by: Adolfo De Unánue <[email protected]>
dssg · Jul 3, 2020 · caf7750 · caf7750
1 parent d9e1e6f
commit caf7750
Show file tree

Hide file tree

Showing 10 changed files with 27 additions and 25 deletions.
diff --git a/requirement/dev.txt b/requirement/dev.txt
@@ -1,6 +1,6 @@
 -r include/build.txt
-bumpversion==0.5.3
-mkdocs==1.0.4
-pymdown-extensions==6.0
-mkdocs-material==4.2.0
-black==19.3b0
+bumpversion==0.6.0
+mkdocs==1.1.2
+pymdown-extensions==7.1
+mkdocs-material==5.4.0
+black==19.10b0
diff --git a/requirement/include/lint.txt b/requirement/include/lint.txt
@@ -1 +1 @@
-flake8==3.7.7
+flake8==3.8.3
diff --git a/requirement/include/test-management.txt b/requirement/include/test-management.txt
@@ -1,3 +1,3 @@
-codecov==2.0.17
+codecov==2.1.7
 coverage>=4.4
-tox==3.9.0
+tox==3.16.1
diff --git a/requirement/main.txt b/requirement/main.txt
@@ -1,12 +1,12 @@
 alembic==1.4.2
-SQLAlchemy==1.3.17
+SQLAlchemy==1.3.18
 PyYAML==5.3.1
 psycopg2-binary==2.8.5
 python-dateutil==2.8.1
-boto3==1.13.21
+boto3==1.14.14
 click==7.1.2
-inflection==0.4.0
-numpy==1.18.4
+inflection==0.5.0
+numpy==1.19.0
 sqlalchemy-postgres-copy==0.5.0
 retrying==1.3.3
 Dickens==1.0.1
@@ -18,12 +18,12 @@ sqlparse==0.3.1
 pebble==4.5.3
 adjustText==0.7.3
 graphviz==0.14
-requests==2.23.0
+requests==2.24.0
 
-scipy==1.3.1
-scikit-learn==0.20.3
-matplotlib==3.2.1
-pandas==1.0.4
+scipy==1.5.0
+scikit-learn==0.23.1
+matplotlib==3.2.2
+pandas==1.0.5
 seaborn==0.10.1
 ohio==0.5.0
 

diff --git a/requirement/test.txt b/requirement/test.txt
@@ -5,7 +5,7 @@ csvkit==1.0.5
 factory_boy==2.12.0
 testing.postgresql==1.3.0
 pytest==5.4.3 #<4.0.0 # pyup: ignore
-pytest-cov==2.9.0
+pytest-cov==2.10.0
 moto==1.3.14
 fakeredis==1.4.1
-hypothesis==5.16.0
+hypothesis==5.19.0
diff --git a/src/tests/catwalk_tests/test_estimators.py b/src/tests/catwalk_tests/test_estimators.py
@@ -69,7 +69,7 @@ def test_dsapp_lr(data):
 
     minmax_scaler = preprocessing.MinMaxScaler()
     dsapp_cutoff = CutOff()
-    lr = linear_model.LogisticRegression()
+    lr = linear_model.LogisticRegression(solver='lbfgs')
 
     pipeline = Pipeline(
         [("minmax_scaler", minmax_scaler), ("dsapp_cutoff", dsapp_cutoff), ("lr", lr)]

diff --git a/src/triage/component/catwalk/estimators/classifiers.py b/src/triage/component/catwalk/estimators/classifiers.py
@@ -26,7 +26,7 @@ def __init__(
         intercept_scaling=1,
         class_weight=None,
         random_state=None,
-        solver="liblinear",
+        solver="lbfgs",
         max_iter=100,
         multi_class="ovr",
         verbose=0,

diff --git a/src/triage/component/catwalk/estimators/transformers.py b/src/triage/component/catwalk/estimators/transformers.py
@@ -45,9 +45,11 @@ def __init__(self, feature_range=(0, 1), copy=True):
         self.feature_range = feature_range
         self.copy = copy
 
+
     def fit(self, X, y=None):
         return self
 
+
     def transform(self, X):
         feature_range = self.feature_range
 
@@ -58,8 +60,8 @@ def transform(self, X):
 
         if np.any(X > feature_range[1]) or np.any(X < feature_range[0]):
             warnings.warn(
-                "You got data that are out of the range: {}".format(
-                    feature_range)
+                f"You got data that are out of the range: {feature_range}. "
+                "It will cutoff to fit in that range."
             )
 
         X[X > feature_range[1]] = feature_range[1]

diff --git a/src/triage/component/catwalk/feature_importances.py b/src/triage/component/catwalk/feature_importances.py
@@ -23,7 +23,7 @@ def _ad_hoc_feature_importances(model):
     """
     feature_importances = None
 
-    if (isinstance(model, (sklearn.linear_model.logistic.LogisticRegression)) or 
+    if (isinstance(model, (sklearn.linear_model.LogisticRegression)) or
         isinstance(model, (ScaledLogisticRegression))):
         coef_odds_ratio = np.exp(model.coef_)
         # intercept_odds_ratio = np.exp(model.intercept_[:,np.newaxis])

diff --git a/src/triage/component/catwalk/storage.py b/src/triage/component/catwalk/storage.py
@@ -13,7 +13,7 @@
 import s3fs
 import wrapt
 import yaml
-from sklearn.externals import joblib
+import joblib
 
 from triage.component.results_schema import (
     TestEvaluation,