Refactor Python tests.

Jenkinsfile

@@ -96,11 +96,11 @@ def buildPlatformCmake(buildName, conf, nodeReq, dockerTarget) {
                 # Test the wheel for compatibility on a barebones CPU container
                 ${dockerRun} release ${dockerArgs} bash -c " \
                     pip install --user python-package/dist/xgboost-*-none-any.whl && \
-                    python -m nose -v tests/python"
+		    pytest -v --fulltrace -s tests/python
                 # Test the wheel for compatibility on CUDA 10.0 container
                 ${dockerRun} gpu --build-arg CUDA_VERSION=10.0 bash -c " \
                     pip install --user python-package/dist/xgboost-*-none-any.whl && \
-                    python -m nose -v --eval-attr='(not slow) and (not mgpu)' tests/python-gpu"
+		    pytest -v -s --fulltrace -m "(not mgpu) and (not slow)" tests/python-gpu
                 """
             }
         }

appveyor.yml

@@ -44,7 +44,7 @@ install:
     - set DO_PYTHON=off
     - if /i "%target%" == "mingw" set DO_PYTHON=on
     - if /i "%target%_%ver%_%configuration%" == "msvc_2015_Release" set DO_PYTHON=on
-    - if /i "%DO_PYTHON%" == "on" conda install -y numpy scipy pandas matplotlib nose scikit-learn graphviz python-graphviz
+    - if /i "%DO_PYTHON%" == "on" conda install -y numpy scipy pandas matplotlib pytest scikit-learn graphviz python-graphviz
     # R: based on https://github.com/krlmlr/r-appveyor
     - ps: |
         if($env:target -eq 'rmingw' -or $env:target -eq 'rmsvc') {
@@ -96,7 +96,7 @@ build_script:
 
 test_script:
     - cd %APPVEYOR_BUILD_FOLDER%
-    - if /i "%DO_PYTHON%" == "on" python -m nose tests/python
+    - if /i "%DO_PYTHON%" == "on" python -m pytest tests/python
     # mingw R package: run the R check (which includes unit tests), and also keep the built binary package
     - if /i "%target%" == "rmingw" (
         set _R_CHECK_CRAN_INCOMING_=FALSE&&

tests/ci_build/Dockerfile.gpu

@@ -37,7 +37,7 @@ ENV CPP=/opt/rh/devtoolset-2/root/usr/bin/cpp
 
 # Install Python packages
 RUN \
-    pip install numpy nose scipy scikit-learn wheel
+    pip install numpy pytest scipy scikit-learn wheel
 
 ENV GOSU_VERSION 1.10
 

tests/ci_build/Dockerfile.release

@@ -15,8 +15,8 @@ ENV PATH=/opt/python/bin:$PATH
 
 # Install Python packages
 RUN \
-    conda install numpy scipy pandas matplotlib nose scikit-learn && \
-    pip install nose wheel auditwheel graphviz
+    conda install numpy scipy pandas matplotlib pytest scikit-learn && \
+    pip install pytest wheel auditwheel graphviz
 
 ENV GOSU_VERSION 1.10
 

tests/ci_build/test_gpu.sh

@@ -4,6 +4,5 @@ set -e
 cd python-package
 python setup.py install --user
 cd ..
-python -m nose -v --eval-attr='(not slow) and (not mgpu)' tests/python-gpu/
+pytest -v -s --fulltrace -m "(not mgpu) and (not slow)" tests/python-gpu
 ./testxgboost --gtest_filter=-*.MGPU_*
-

tests/ci_build/test_mgpu.sh

@@ -4,5 +4,5 @@ set -e
 cd python-package
 python setup.py install --user
 cd ..
-python -m nose -v --eval-attr='(not slow) and mgpu' tests/python-gpu/
+pytest -v -s --fulltrace -m "(not slow) and mgpu" tests/python-gpu
 ./testxgboost --gtest_filter=*.MGPU_*

tests/python-gpu/test_gpu_linear.py

@@ -1,18 +1,19 @@
 import sys
+import pytest
+import unittest
 
 sys.path.append('tests/python/')
 import test_linear
 import testing as tm
-import unittest
 
 
 class TestGPULinear(unittest.TestCase):
 
     datasets = ["Boston", "Digits", "Cancer", "Sparse regression",
                 "Boston External Memory"]
-
+
+    @pytest.mark.skipif(**tm.no_sklearn())
     def test_gpu_coordinate(self):
-        tm._skip_if_no_sklearn()
         variable_param = {
             'booster': ['gblinear'],
             'updater': ['coord_descent'],

tests/python-gpu/test_gpu_prediction.py

@@ -1,15 +1,14 @@
 from __future__ import print_function
 
 import numpy as np
-import sys
 import unittest
 import xgboost as xgb
-from nose.plugins.attrib import attr
+import pytest
 
 rng = np.random.RandomState(1994)
 
 
-@attr('gpu')
+@pytest.mark.gpu
 class TestGPUPredict(unittest.TestCase):
     def test_predict(self):
         iterations = 10
@@ -18,17 +17,21 @@ def test_predict(self):
         test_num_cols = [10, 50, 500]
         for num_rows in test_num_rows:
             for num_cols in test_num_cols:
-                dtrain = xgb.DMatrix(np.random.randn(num_rows, num_cols), label=[0, 1] * int(num_rows / 2))
-                dval = xgb.DMatrix(np.random.randn(num_rows, num_cols), label=[0, 1] * int(num_rows / 2))
-                dtest = xgb.DMatrix(np.random.randn(num_rows, num_cols), label=[0, 1] * int(num_rows / 2))
+                dtrain = xgb.DMatrix(np.random.randn(num_rows, num_cols),
+                                     label=[0, 1] * int(num_rows / 2))
+                dval = xgb.DMatrix(np.random.randn(num_rows, num_cols),
+                                   label=[0, 1] * int(num_rows / 2))
+                dtest = xgb.DMatrix(np.random.randn(num_rows, num_cols),
+                                    label=[0, 1] * int(num_rows / 2))
                 watchlist = [(dtrain, 'train'), (dval, 'validation')]
                 res = {}
                 param = {
                     "objective": "binary:logistic",
                     "predictor": "gpu_predictor",
                     'eval_metric': 'auc',
                 }
-                bst = xgb.train(param, dtrain, iterations, evals=watchlist, evals_result=res)
+                bst = xgb.train(param, dtrain, iterations, evals=watchlist,
+                                evals_result=res)
                 assert self.non_decreasing(res["train"]["auc"])
                 gpu_pred_train = bst.predict(dtrain, output_margin=True)
                 gpu_pred_test = bst.predict(dtest, output_margin=True)
@@ -39,21 +42,26 @@ def test_predict(self):
                 cpu_pred_train = bst_cpu.predict(dtrain, output_margin=True)
                 cpu_pred_test = bst_cpu.predict(dtest, output_margin=True)
                 cpu_pred_val = bst_cpu.predict(dval, output_margin=True)
-                np.testing.assert_allclose(cpu_pred_train, gpu_pred_train, rtol=1e-5)
-                np.testing.assert_allclose(cpu_pred_val, gpu_pred_val, rtol=1e-5)
-                np.testing.assert_allclose(cpu_pred_test, gpu_pred_test, rtol=1e-5)
+                np.testing.assert_allclose(cpu_pred_train, gpu_pred_train,
+                                           rtol=1e-5)
+                np.testing.assert_allclose(cpu_pred_val, gpu_pred_val,
+                                           rtol=1e-5)
+                np.testing.assert_allclose(cpu_pred_test, gpu_pred_test,
+                                           rtol=1e-5)
 
     def non_decreasing(self, L):
         return all((x - y) < 0.001 for x, y in zip(L, L[1:]))
 
-    # Test case for a bug where multiple batch predictions made on a test set produce incorrect results
+    # Test case for a bug where multiple batch predictions made on a
+    # test set produce incorrect results
     def test_multi_predict(self):
         from sklearn.datasets import make_regression
         from sklearn.model_selection import train_test_split
 
         n = 1000
         X, y = make_regression(n, random_state=rng)
-        X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=123)
+        X_train, X_test, y_train, y_test = train_test_split(X, y,
+                                                            random_state=123)
         dtrain = xgb.DMatrix(X_train, label=y_train)
         dtest = xgb.DMatrix(X_test)
 
@@ -85,8 +93,7 @@ def test_sklearn(self):
         params = {'tree_method': 'gpu_hist',
                   'predictor': 'cpu_predictor',
                   'n_jobs': -1,
-                  'seed': 123
-                  }
+                  'seed': 123}
         m = xgb.XGBRegressor(**params).fit(X_train, y_train)
         cpu_train_score = m.score(X_train, y_train)
         cpu_test_score = m.score(X_test, y_test)

tests/python-gpu/test_gpu_updaters.py

@@ -1,10 +1,9 @@
 import numpy as np
 import sys
 import unittest
-from nose.plugins.attrib import attr
+import pytest
 
 sys.path.append("tests/python")
-import xgboost as xgb
 from regression_test_utilities import run_suite, parameter_combinations, \
     assert_results_non_increasing
 
@@ -45,7 +44,7 @@ def test_gpu_hist(self):
             cpu_results = run_suite(param, select_datasets=datasets)
             assert_gpu_results(cpu_results, gpu_results)
 
-    @attr('mgpu')
+    @pytest.mark.mgpu
     def test_gpu_hist_mgpu(self):
         variable_param = {'n_gpus': [-1], 'max_depth': [2, 10],
                           'max_leaves': [255, 4],
@@ -56,7 +55,7 @@ def test_gpu_hist_mgpu(self):
             gpu_results = run_suite(param, select_datasets=datasets)
             assert_results_non_increasing(gpu_results, 1e-2)
 
-    @attr('mgpu')
+    @pytest.mark.mgpu
     def test_specified_gpu_id_gpu_update(self):
         variable_param = {'n_gpus': [1],
                           'gpu_id': [1],

tests/python-gpu/test_large_sizes.py

@@ -2,12 +2,12 @@
 
 import sys
 import time
+import pytest
 
 sys.path.append("../../tests/python")
 import xgboost as xgb
 import numpy as np
 import unittest
-from nose.plugins.attrib import attr
 
 
 def eprint(*args, **kwargs):
@@ -16,9 +16,11 @@ def eprint(*args, **kwargs):
     print(*args, file=sys.stdout, **kwargs)
     sys.stdout.flush()
 
+
 rng = np.random.RandomState(1994)
 
-# "realistic" size based upon http://stat-computing.org/dataexpo/2009/ , which has been processed to one-hot encode categoricalsxsy
+# "realistic" size based upon http://stat-computing.org/dataexpo/2009/
+# , which has been processed to one-hot encode categoricalsxsy
 cols = 31
 # reduced to fit onto 1 gpu but still be large
 rows3 = 5000  # small
@@ -28,7 +30,7 @@ def eprint(*args, **kwargs):
 rowslist = [rows1, rows2, rows3]
 
 
-@attr('slow')
+@pytest.mark.slow
 class TestGPU(unittest.TestCase):
     def test_large(self):
         for rows in rowslist:
@@ -47,15 +49,8 @@ def test_large(self):
             max_depth = 6
             max_bin = 1024
 
-            # regression test --- hist must be same as exact on all-categorial data
-            ag_param = {'max_depth': max_depth,
-                        'tree_method': 'exact',
-                        'nthread': 0,
-                        'eta': 1,
-                        'silent': 0,
-                        'debug_verbose': 5,
-                        'objective': 'binary:logistic',
-                        'eval_metric': 'auc'}
+            # regression test --- hist must be same as exact on
+            # all-categorial data
             ag_paramb = {'max_depth': max_depth,
                          'tree_method': 'hist',
                          'nthread': 0,

tests/python-gpu/test_monotonic_constraints.py

@@ -1,10 +1,12 @@
 from __future__ import print_function
 
 import numpy as np
+from sklearn.datasets import make_regression
+
 import unittest
+import pytest
+
 import xgboost as xgb
-from nose.plugins.attrib import attr
-from sklearn.datasets import make_regression
 
 rng = np.random.RandomState(1994)
 
@@ -33,7 +35,7 @@ def assert_constraint(constraint, tree_method):
         assert non_increasing(pred)
 
 
-@attr('gpu')
+@pytest.mark.gpu
 class TestMonotonicConstraints(unittest.TestCase):
     def test_exact(self):
         assert_constraint(1, 'exact')

tests/python/test_basic.py

@@ -21,6 +21,8 @@ def captured_output():
     """
     Reassign stdout temporarily in order to test printed statements
     Taken from: https://stackoverflow.com/questions/4219717/how-to-assert-output-with-nosetest-unittest-in-python
+
+    Also works for pytest.
     """
     new_out, new_err = StringIO(), StringIO()
     old_out, old_err = sys.stdout, sys.stderr
@@ -36,15 +38,17 @@ class TestBasic(unittest.TestCase):
     def test_basic(self):
         dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
         dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
-        param = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'binary:logistic'}
+        param = {'max_depth': 2, 'eta': 1, 'silent': 1,
+                 'objective': 'binary:logistic'}
         # specify validations set to watch performance
         watchlist = [(dtest, 'eval'), (dtrain, 'train')]
         num_round = 2
         bst = xgb.train(param, dtrain, num_round, watchlist)
         # this is prediction
         preds = bst.predict(dtest)
         labels = dtest.get_label()
-        err = sum(1 for i in range(len(preds)) if int(preds[i] > 0.5) != labels[i]) / float(len(preds))
+        err = sum(1 for i in range(len(preds))
+                  if int(preds[i] > 0.5) != labels[i]) / float(len(preds))
         # error must be smaller than 10%
         assert err < 0.1
 
@@ -62,7 +66,8 @@ def test_basic(self):
     def test_record_results(self):
         dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')
         dtest = xgb.DMatrix(dpath + 'agaricus.txt.test')
-        param = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'binary:logistic'}
+        param = {'max_depth': 2, 'eta': 1, 'silent': 1,
+                 'objective': 'binary:logistic'}
         # specify validations set to watch performance
         watchlist = [(dtest, 'eval'), (dtrain, 'train')]
         num_round = 2
@@ -86,7 +91,8 @@ def test_multiclass(self):
         # this is prediction
         preds = bst.predict(dtest)
         labels = dtest.get_label()
-        err = sum(1 for i in range(len(preds)) if preds[i] != labels[i]) / float(len(preds))
+        err = sum(1 for i in range(len(preds))
+                  if preds[i] != labels[i]) / float(len(preds))
         # error must be smaller than 10%
         assert err < 0.1
 
@@ -248,7 +254,8 @@ def test_dmatrix_numpy_init(self):
 
     def test_cv(self):
         dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
-        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'binary:logistic'}
+        params = {'max_depth': 2, 'eta': 1, 'silent': 1,
+                  'objective': 'binary:logistic'}
 
         # return np.ndarray
         cv = xgb.cv(params, dm, num_boost_round=10, nfold=10, as_pandas=False)
@@ -257,29 +264,34 @@ def test_cv(self):
 
     def test_cv_no_shuffle(self):
         dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
-        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'binary:logistic'}
+        params = {'max_depth': 2, 'eta': 1, 'silent': 1,
+                  'objective': 'binary:logistic'}
 
         # return np.ndarray
-        cv = xgb.cv(params, dm, num_boost_round=10, shuffle=False, nfold=10, as_pandas=False)
+        cv = xgb.cv(params, dm, num_boost_round=10, shuffle=False, nfold=10,
+                    as_pandas=False)
         assert isinstance(cv, dict)
         assert len(cv) == (4)
 
     def test_cv_explicit_fold_indices(self):
         dm = xgb.DMatrix(dpath + 'agaricus.txt.train')
-        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'binary:logistic'}
+        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective':
+                  'binary:logistic'}
         folds = [
             # Train        Test
             ([1, 3], [5, 8]),
             ([7, 9], [23, 43]),
         ]
 
         # return np.ndarray
-        cv = xgb.cv(params, dm, num_boost_round=10, folds=folds, as_pandas=False)
+        cv = xgb.cv(params, dm, num_boost_round=10, folds=folds,
+                    as_pandas=False)
         assert isinstance(cv, dict)
         assert len(cv) == (4)
 
     def test_cv_explicit_fold_indices_labels(self):
-        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective': 'reg:linear'}
+        params = {'max_depth': 2, 'eta': 1, 'silent': 1, 'objective':
+                  'reg:linear'}
         N = 100
         F = 3
         dm = xgb.DMatrix(data=np.random.randn(N, F), label=np.arange(N))
@@ -300,7 +312,9 @@ def cb(cbackenv):
                 as_pandas=False
             )
             output = out.getvalue().strip()
-        assert output == '[array([5., 8.], dtype=float32), array([23., 43., 11.], dtype=float32)]'
+        solution = ('[array([5., 8.], dtype=float32), array([23., 43., 11.],' +
+                    ' dtype=float32)]')
+        assert output == solution
 
     def test_get_info(self):
         dtrain = xgb.DMatrix(dpath + 'agaricus.txt.train')