Merge pull request #55 from KrishnaswamyLab/dev

Bugfixes to address warnings from sklearn, anndata changes

.travis.yml

@@ -2,7 +2,8 @@ language: python
 python:
   - '3.5'
   - '3.6'
-  - 3.7-dev
+  - '3.7'
+  - '3.8'
 cache:
   - pip
   - apt
@@ -29,4 +30,3 @@ deploy:
     branch: master
 after_success:
   - coveralls
-

graphtools/graphs.py

@@ -288,7 +288,7 @@ def _check_duplicates(self, distances, indices):
                     "Detected zero distance between {} pairs of samples. "
                     "Consider removing duplicates to avoid errors in "
                     "downstream processing.".format(
-                        np.sum(np.sum(distances[:, 1:] == 0))
+                        np.sum(np.sum(distances[:, 1:] == 0)) // 2
                     ),
                     RuntimeWarning,
                 )
@@ -408,7 +408,7 @@ def build_kernel_to_data(
                     search_knn = min(search_knn * self.search_multiplier, knn_max)
                 if search_knn > self.data_nu.shape[0] / 2:
                     knn_tree = NearestNeighbors(
-                        search_knn, algorithm="brute", n_jobs=self.n_jobs
+                        n_neighbors=search_knn, algorithm="brute", n_jobs=self.n_jobs
                     ).fit(self.data_nu)
                 if len(update_idx) > 0:
                     if search_knn == knn_max:

graphtools/version.py

@@ -1 +1 @@
-__version__ = "1.5.1"
+__version__ = "1.5.2"

test/test_estimator.py

@@ -95,7 +95,7 @@ def test_pca():
 def test_anndata_input():
     X = np.random.normal(0, 1, (10, 2))
     E = Estimator(verbose=0)
-    E.fit(X)
+    E.fit(X.astype(np.float32))
     E2 = Estimator(verbose=0)
     E2.fit(anndata.AnnData(X))
     np.testing.assert_allclose(

test/test_knn.py

@@ -51,7 +51,7 @@ def test_duplicate_data():
         RuntimeWarning,
         r"Detected zero distance between samples ([0-9and,\s]*). Consider removing duplicates to avoid errors in downstream processing.",
     ):
-        build_graph(np.vstack([data, data[:10]]), n_pca=20, decay=10, thresh=1e-4)
+        build_graph(np.vstack([data, data[:9]]), n_pca=20, decay=10, thresh=1e-4)
 
 
 def test_duplicate_data_many():