CI: Test that sort order is enforced. (#314)

.circleci/config.yml

@@ -16,17 +16,27 @@ jobs:
       - run:
           command: |
             source activate ${ENV_NAME}
-            pytest --cov=dask_ml tests
+            mkdir test-reports
+            pytest --cov=dask_ml --junitxml=test-reports/junit.xml tests
       - run:
           command: |
             source activate ${ENV_NAME}
+            echo "[flake8]"
             flake8
+            echo "[black]"
             black --check .
+            echo "[isort]"
+            isort --recursive --check-only .
+            echo "[codecov]"
             codecov
       - save_cache:
           key: deps-{{ .Branch }}-3.6-{{ checksum "ci/environment-3.6.yml" }}
           paths:
             - "/opt/conda/envs/${ENV_NAME}/"
+      - store_test_results:
+          path: test-reports
+      - store_artifacts:
+          path: test-reports
 
   # Python 2.7 build
   py27:
@@ -45,14 +55,18 @@ jobs:
       - run:
           command: |
             source activate ${ENV_NAME}
-            pytest tests
+            pytest --junitxml=test-reports/junit.xml tests
       - run:
           command: |
             source activate ${ENV_NAME}
       - save_cache:
           key: deps-{{ .Branch }}-2.7-{{ checksum "ci/environment-2.7.yml" }}
           paths:
             - "/opt/conda/envs/${ENV_NAME}/"
+      - store_test_results:
+          path: test-reports
+      - store_artifacts:
+          path: test-reports
 
   # scikit-learn dev build
   sklearn_dev:
@@ -77,11 +91,15 @@ jobs:
       - run:
           command: |
             source activate ${ENV_NAME}
-            pytest tests
+            pytest --junitxml=test-reports/junit.xml tests
       - save_cache:
           key: deps-{{ .Branch }}-sklearn_dev-{{ checksum "ci/environment-3.6.yml" }}
           paths:
             - "/opt/conda/envs/${ENV_NAME}/"
+      - store_test_results:
+          path: test-reports
+      - store_artifacts:
+          path: test-reports
 
 workflows:
   version: 2

benchmarks/k_means_kdd.py

@@ -12,15 +12,14 @@
 import dask.dataframe as dd
 import pandas as pd
 import requests
+import sklearn.cluster as sk
 from distributed import Client
+from sklearn.datasets import get_data_home
 
 import s3fs
-import sklearn.cluster as sk
 from dask_ml.cluster import KMeans
 from dask_ml.utils import _timer
 
-from sklearn.datasets import get_data_home
-
 from .base import make_parser
 
 logger = logging.getLogger(__name__)

benchmarks/kmeans_airline.py

@@ -6,7 +6,6 @@
 import sys
 from timeit import default_timer as tic
 
-import coloredlogs
 import dask.array as da
 import dask.dataframe as dd
 import pandas as pd
@@ -17,7 +16,6 @@
 from dask_ml.utils import _timer
 
 logger = logging.getLogger(__name__)
-coloredlogs.install()
 
 
 def parse_args(args=None):

ci/environment-3.6.yml

@@ -10,6 +10,7 @@ dependencies:
   - flake8
   - heapdict
   - ipython
+  - isort
   - multipledispatch
   - nbsphinx
   - nomkl

ci/install-circle.sh

@@ -3,9 +3,9 @@ set -xe
 apt-get update; apt-get install -y gcc g++
 conda config --set always_yes true --set changeps1 false --set quiet true
 conda update -q conda
-conda install conda-build anaconda-client --yes
+conda install conda-build anaconda-client --yes --quiet
 conda config --add channels conda-forge
-conda env create -f ci/environment-${PYTHON}.yml --name=${ENV_NAME}
+conda env create -f ci/environment-${PYTHON}.yml --name=${ENV_NAME} --quiet
 conda env list
 source activate ${ENV_NAME}
 pip install --no-deps --quiet -e .

dask_ml/_compat.py

@@ -2,7 +2,6 @@
 
 import dask
 import packaging.version
-
 import sklearn
 
 SK_VERSION = packaging.version.parse(sklearn.__version__)

dask_ml/_partial.py

@@ -8,11 +8,10 @@
 import dask
 import numpy as np
 import six
+import sklearn.utils
 from dask.delayed import Delayed
 from toolz import partial
 
-import sklearn.utils
-
 from ._utils import copy_learned_attributes
 
 logger = logging.getLogger(__name__)

dask_ml/cluster/k_means.py

@@ -7,7 +7,6 @@
 import numpy as np
 import pandas as pd
 from dask import compute
-
 from sklearn.base import BaseEstimator, TransformerMixin
 from sklearn.cluster import k_means_ as sk_k_means
 from sklearn.utils.extmath import squared_norm
@@ -19,7 +18,6 @@
     pairwise_distances_argmin_min,
 )
 from ..utils import _timed, _timer, check_array, row_norms
-
 from ._k_means import _centers_dense
 
 logger = logging.getLogger(__name__)

dask_ml/cluster/spectral.py

@@ -6,10 +6,9 @@
 import dask.array as da
 import numpy as np
 import six
+import sklearn.cluster
 from dask import delayed
 from scipy.linalg import pinv, svd
-
-import sklearn.cluster
 from sklearn.base import BaseEstimator, ClusterMixin
 from sklearn.utils import check_random_state
 

dask_ml/datasets.py

@@ -3,11 +3,11 @@
 import dask
 import dask.array as da
 import numpy as np
-
-import dask_ml.utils
 import sklearn.datasets
 import sklearn.utils
 
+import dask_ml.utils
+
 
 def _check_axis_partitioning(chunks, n_features):
     c = chunks[1][0]

dask_ml/decomposition/pca.py

@@ -2,7 +2,6 @@
 import numpy as np
 import scipy.sparse as sp
 from dask import compute
-
 from sklearn.decomposition.base import _BasePCA
 from sklearn.utils.extmath import fast_logdet
 from sklearn.utils.validation import check_is_fitted, check_random_state

dask_ml/decomposition/truncated_svd.py

@@ -1,6 +1,5 @@
 import dask.array as da
 from dask import compute
-
 from sklearn.base import BaseEstimator, TransformerMixin
 
 from ..utils import svd_flip

dask_ml/linear_model/glm.py

@@ -14,7 +14,7 @@
 )
 from sklearn.base import BaseEstimator
 
-# register multipledispatch
+# Register multiple dispatch
 from . import utils  # noqa
 from ..utils import check_array
 

dask_ml/metrics/pairwise.py

@@ -8,7 +8,6 @@
 import packaging.version
 from dask import delayed
 from dask.array.random import doc_wraps
-
 from sklearn import metrics
 from sklearn.metrics.pairwise import KERNEL_PARAMS
 

dask_ml/metrics/regression.py

@@ -1,8 +1,7 @@
 import dask.array as da
 import six
-from dask.array.random import doc_wraps
-
 import sklearn.metrics
+from dask.array.random import doc_wraps
 
 
 def _check_sample_weight(sample_weight):

dask_ml/metrics/scorer.py

@@ -1,5 +1,4 @@
 import six
-
 from sklearn.metrics import make_scorer
 from sklearn.metrics.scorer import check_scoring as sklearn_check_scoring
 

dask_ml/model_selection/_normalize.py

@@ -2,7 +2,6 @@
 
 import numpy as np
 from dask.base import normalize_token
-
 from sklearn.base import BaseEstimator
 from sklearn.model_selection._split import (
     BaseShuffleSplit,

dask_ml/model_selection/_search.py

@@ -12,7 +12,6 @@
 from dask.base import tokenize
 from dask.delayed import delayed
 from dask.utils import derived_from
-
 from sklearn import model_selection
 from sklearn.base import BaseEstimator, MetaEstimatorMixin, clone, is_classifier
 from sklearn.exceptions import NotFittedError

dask_ml/model_selection/_split.py

@@ -6,16 +6,16 @@
 import dask
 import dask.array as da
 import numpy as np
-
 import sklearn.model_selection as ms
-from dask_ml.utils import check_array
 from sklearn.model_selection._split import (
     BaseCrossValidator,
     _validate_shuffle_split,
     _validate_shuffle_split_init,
 )
 from sklearn.utils import check_random_state
 
+from dask_ml.utils import check_array
+
 
 def _check_blockwise(blockwise):
     if blockwise not in {True, False}:

dask_ml/model_selection/methods.py

@@ -10,12 +10,11 @@
 from dask.base import normalize_token
 from scipy import sparse
 from scipy.stats import rankdata
-from toolz import pluck
-
 from sklearn.exceptions import FitFailedWarning
 from sklearn.pipeline import FeatureUnion, Pipeline
 from sklearn.utils import safe_indexing
 from sklearn.utils.validation import _is_arraylike, check_consistent_length
+from toolz import pluck
 
 from .utils import copy_estimator
 

dask_ml/model_selection/utils.py

@@ -5,7 +5,6 @@
 import dask.array as da
 from dask.base import tokenize
 from dask.delayed import Delayed, delayed
-
 from sklearn.utils.validation import _is_arraylike, indexable
 
 if LooseVersion(dask.__version__) > "0.15.4":

dask_ml/model_selection/utils_test.py

@@ -2,7 +2,6 @@
 from functools import wraps
 
 import numpy as np
-
 from sklearn.base import BaseEstimator, ClassifierMixin
 from sklearn.utils.validation import _num_samples, check_array
 

dask_ml/naive_bayes.py

@@ -1,7 +1,6 @@
 import dask.array as da
 import numpy as np
 from dask import delayed
-
 from sklearn import naive_bayes as _naive_bayes
 from sklearn.base import BaseEstimator
 

dask_ml/preprocessing/data.py

@@ -11,12 +11,12 @@
 from dask import compute
 from pandas.api.types import is_categorical_dtype
 from scipy import stats
-
-from dask_ml.utils import check_array, handle_zeros_in_scale
 from sklearn.base import BaseEstimator, TransformerMixin
 from sklearn.preprocessing import data as skdata
 from sklearn.utils.validation import check_is_fitted, check_random_state
 
+from dask_ml.utils import check_array, handle_zeros_in_scale
+
 _PANDAS_VERSION = LooseVersion(pd.__version__)
 _HAS_CTD = _PANDAS_VERSION >= "0.21.0"