added unit tests for anndata raw

Author: davidsebfischer

batchglm/data.py

@@ -67,21 +67,26 @@ def xarray_from_data(
     :param dims: tuple or list with two strings. Specifies the names of the xarray dimensions.
     :return: xr.DataArray of shape `dims`
     """
-    if anndata is not None and isinstance(data, anndata.AnnData):
+    if anndata is not None and (isinstance(data, anndata.AnnData) or isinstance(data, anndata.base.Raw)):
+        # Anndata.raw does not have obs_names.
+        if isinstance(data, anndata.AnnData):
+            obs_names = np.asarray(data.obs_names)
+        else:
+            obs_names = ["obs_" + str(i) for i in range(data.X.shape[0])]
+
         if scipy.sparse.issparse(data.X):
             # X = _sparse_to_xarray(data.X, dims=dims)
             # X.coords[dims[0]] = np.asarray(data.obs_names)
             # X.coords[dims[1]] = np.asarray(data.var_names)
             X = SparseXArrayDataSet(
                 X=data.X,
-                obs_names=np.asarray(data.obs_names),
+                obs_names=np.asarray(obs_names),
                 feature_names=np.asarray(data.var_names),
                 dims=dims
             )
         else:
-            X = data.X
-            X = xr.DataArray(X, dims=dims, coords={
-                dims[0]: np.asarray(data.obs_names),
+            X = xr.DataArray(data.X, dims=dims, coords={
+                dims[0]: np.asarray(obs_names),
                 dims[1]: np.asarray(data.var_names),
             })
     elif isinstance(data, xr.Dataset):

batchglm/unit_test/base_glm/test_data_types_glm.py

@@ -91,36 +91,41 @@ def basic_test(
         estimator = self.get_estimator(input_data=input_data)
         return estimator.test_estimation()
 
-    def _test_numpy_dense(self):
-        return self.basic_test(
-            data=self.sim.X,
-            design_loc=self.sim.design_loc,
-            design_scale=self.sim.design_scale
-        )
+    def _test_numpy(self, sparse):
+        if sparse:
+            X = scipy.sparse.csr_matrix(self.sim.X)
 
-    def _test_scipy_sparse(self):
-        return self.basic_test(
-            data=scipy.sparse.csr_matrix(self.sim.X),
+        success = self.basic_test(
+            data=X,
             design_loc=self.sim.design_loc,
             design_scale=self.sim.design_scale
         )
+        assert success, "_test_anndata with sparse=%s did not work" % sparse
 
-    def _test_anndata_dense(self):
+    def _test_anndata(self, sparse):
         adata = self.sim.data_to_anndata()
-        return self.basic_test(
+        if sparse:
+            adata.X = scipy.sparse.csr_matrix(adata.X)
+
+        success = self.basic_test(
             data=adata,
             design_loc=self.sim.design_loc,
             design_scale=self.sim.design_scale
         )
+        assert success, "_test_anndata with sparse=%s did not work" % sparse
 
-    def _test_anndata_sparse(self):
+    def _test_anndata_raw(self, sparse):
         adata = self.sim.data_to_anndata()
-        adata.X = scipy.sparse.csr_matrix(adata.X)
-        return self.basic_test(
-            data=adata,
+        if sparse:
+            adata.X = scipy.sparse.csr_matrix(adata.X)
+
+        adata.raw = adata
+        success = self.basic_test(
+            data=adata.raw,
             design_loc=self.sim.design_loc,
             design_scale=self.sim.design_scale
         )
+        assert success, "_test_anndata with sparse=%s did not work" % sparse
 
 
 if __name__ == '__main__':

batchglm/unit_test/glm_all/test_data_types_glm_all.py

@@ -99,18 +99,6 @@ def get_estimator(
             noise_model=self.noise_model
         )
 
-    def _test_standard(self):
-        self.simulate()
-        logger.debug("* Running tests on numpy/scipy")
-        self._test_numpy_dense()
-        self._test_scipy_sparse()
-
-    def _test_anndata(self):
-        self.simulate()
-        logger.debug("* Running tests on anndata")
-        self._test_anndata_dense()
-        self._test_anndata_sparse()
-
 
 class Test_DataTypes_GLM_NB(
     Test_DataTypes_GLM_ALL,
@@ -120,22 +108,31 @@ class Test_DataTypes_GLM_NB(
     Test whether training graphs work for negative binomial noise.
     """
 
-    def test_standard_nb(self):
+    def test_standard(self):
         logging.getLogger("tensorflow").setLevel(logging.ERROR)
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logger.error("Test_DataTypes_GLM_NB.test_standard_nb()")
 
         self.noise_model = "nb"
-        self._test_standard()
+        self.simulate()
+        self._test_numpy(sparse=False)
+        self._test_numpy(sparse=True)
+
+        return True
 
-    def test_anndata_nb(self):
+    def test_anndata(self):
         logging.getLogger("tensorflow").setLevel(logging.ERROR)
         logging.getLogger("batchglm").setLevel(logging.WARNING)
         logger.error("Test_DataTypes_GLM_NB.test_anndata_nb()")
 
         self.noise_model = "nb"
-        self._test_anndata()
+        self.simulate()
+        self._test_anndata(sparse=False)
+        self._test_anndata(sparse=True)
+        self._test_anndata_raw(sparse=False)
+        self._test_anndata_raw(sparse=True)
 
+        return True
 
 if __name__ == '__main__':
     unittest.main()