fixed loom import export

Author: falexwolf

anndata/readwrite/read.py

@@ -1,6 +1,6 @@
 from pathlib import Path
 from typing import Union, Optional, Iterable, Generator, Iterator
-
+from scipy.sparse import issparse
 import numpy as np
 
 from ..base import AnnData

anndata/readwrite/write.py

@@ -3,11 +3,10 @@
 from collections import Mapping
 from pathlib import Path
 from typing import Union
-
 import pandas as pd
 import numpy as np
-from scipy import sparse
 from scipy.sparse import issparse
+import logging as logg
 
 from ..base import AnnData
 from .. import h5py
@@ -58,18 +57,18 @@ def write_csvs(dirname, adata, skip_data=True, sep=','):
 
 
 def write_loom(filename: Union[Path, str], adata: AnnData):
-    from loompy import create
+    filename = str(filename)  # allow passing Path object
     row_attrs = adata.var.to_dict('list')
     row_attrs['var_names'] = adata.var_names.values
     col_attrs = adata.obs.to_dict('list')
     col_attrs['obs_names'] = adata.obs_names.values
-    with create(
-        str(filename),  # allow passing Path objects
-        adata.X.T,
-        row_attrs=row_attrs,
-        col_attrs=col_attrs
-    ) as ds:
-        pass
+    X = adata.X.T
+    if issparse(X):
+        logg.info(
+            '... writing to \'.loom\' file densifies sparse matrix')
+        X = X.toarray()
+    from loompy import create
+    create(filename, X, row_attrs=row_attrs, col_attrs=col_attrs)
 
 
 def _write_h5ad(filename: Union[Path, str], adata: AnnData, **kwargs):

anndata/tests/readwrite.py

@@ -60,17 +60,17 @@ def test_readwrite_h5ad():
         assert adata.obs['oanno1'].cat.categories.tolist() == ['cat1', 'cat2']
 
 
-# def test_readwrite_loom():
-#     for typ in [np.array, csr_matrix]:
-#         X = typ(X_list)
-#         adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
-#         adata.write_loom('./test.loom')
-#         adata = ad.read_loom('./test.loom')
-#         if isinstance(X, np.ndarray):
-#             assert np.allclose(adata.X, X)
-#         else:
-#             # TODO: this should not be necessary
-#             assert np.allclose(adata.X, X.toarray())
+def test_readwrite_loom():
+    for typ in [np.array, csr_matrix]:
+        X = typ(X_list)
+        adata = ad.AnnData(X, obs=obs_dict, var=var_dict, uns=uns_dict)
+        adata.write_loom('./test.loom')
+        adata = ad.read_loom('./test.loom')
+        if isinstance(X, np.ndarray):
+            assert np.allclose(adata.X, X)
+        else:
+            # TODO: this should not be necessary
+            assert np.allclose(adata.X, X.toarray())
 
 
 def test_read_csv():