Added finding quantiles on GPU. (#3393)

* Added finding quantiles on GPU.

- this includes datasets where weights are assigned to data rows
- as the quantiles found by the new algorithm are not the same
  as those found by the old one, test thresholds in
    tests/python-gpu/test_gpu_updaters.py have been adjusted.

* Adjustments and improved testing for finding quantiles on the GPU.

- added C++ tests for the DeviceSketch() function
- reduced one of the thresholds in test_gpu_updaters.py
- adjusted the cuts found by the find_cuts_k kernel

tests/python-gpu/test_gpu_linear.py

@@ -7,12 +7,26 @@ import unittest
 
 
 class TestGPULinear(unittest.TestCase):
+
+    datasets = ["Boston", "Digits", "Cancer", "Sparse regression",
+                "Boston External Memory"]
+    
     def test_gpu_coordinate(self):
         tm._skip_if_no_sklearn()
-        variable_param = {'booster': ['gblinear'], 'updater': ['coord_descent'], 'eta': [0.5],
-                          'top_k': [10], 'tolerance': [1e-5], 'nthread': [2], 'alpha': [.005, .1], 'lambda': [0.005],
-                          'coordinate_selection': ['cyclic', 'random', 'greedy'], 'n_gpus': [-1]}
+        variable_param = {
+            'booster': ['gblinear'],
+            'updater': ['coord_descent'],
+            'eta': [0.5],
+            'top_k': [10],
+            'tolerance': [1e-5],
+            'nthread': [2],
+            'alpha': [.005, .1],
+            'lambda': [0.005],
+            'coordinate_selection': ['cyclic', 'random', 'greedy'],
+            'n_gpus': [-1]
+        }
         for param in test_linear.parameter_combinations(variable_param):
-            results = test_linear.run_suite(param, 200, None, scale_features=True)
+            results = test_linear.run_suite(
+                param, 200, self.datasets, scale_features=True)
             test_linear.assert_regression_result(results, 1e-2)
             test_linear.assert_classification_result(results)

tests/python-gpu/test_gpu_updaters.py

@@ -11,11 +11,10 @@ from regression_test_utilities import run_suite, parameter_combinations, \
 def assert_gpu_results(cpu_results, gpu_results):
     for cpu_res, gpu_res in zip(cpu_results, gpu_results):
         # Check final eval result roughly equivalent
-        assert np.allclose(cpu_res["eval"][-1], gpu_res["eval"][-1], 1e-3, 1e-2)
-
-
-datasets = ["Boston", "Cancer", "Digits", "Sparse regression"]
+        assert np.allclose(cpu_res["eval"][-1], gpu_res["eval"][-1], 1e-2, 1e-2)
 
+datasets = ["Boston", "Cancer", "Digits", "Sparse regression",
+            "Sparse regression with weights"]
 
 class TestGPU(unittest.TestCase):
     def test_gpu_exact(self):