Do dynamic imports of graspologic for performance

src/pixelator/graph/community_detection.py

@@ -5,17 +5,13 @@
 
 import logging
 from copy import copy
-from dataclasses import dataclass
 from pathlib import Path
-from queue import Queue
-from time import time
-from typing import Optional, Tuple
+from typing import Tuple
 
 import networkx as nx
 import numpy as np
 import pandas as pd
 import polars as pl
-from graspologic.partition import leiden
 
 from pixelator.graph.constants import (
     LEIDEN_RESOLUTION,
@@ -25,7 +21,6 @@
     edgelist_metrics,
     map_upis_to_components,
     split_remaining_and_removed_edgelist,
-    update_edgelist_membership,
 )
 from pixelator.report.models.graph import GraphSampleReport
 from pixelator.types import PathType
@@ -297,6 +292,9 @@ def recover_technical_multiplets(
         edgelist.shape[0],
     )
 
+    # Import here since the import is very slow and expensive
+    from graspologic.partition import leiden
+
     def id_generator(start=0):
         next_id = start
         while True:

src/pixelator/pixeldataset/utils.py

@@ -13,7 +13,6 @@
 import numpy as np
 import pandas as pd
 from anndata import AnnData, ImplicitModificationWarning, read_h5ad
-from graspologic.partition import leiden
 
 from pixelator.graph import components_metrics
 from pixelator.graph.constants import LEIDEN_RESOLUTION, RELATIVE_ANNOTATE_RESOLUTION
@@ -222,6 +221,9 @@ def write_anndata(adata: AnnData, filename: PathType) -> None:
 def _compute_sub_communities(
     component_edgelist: pd.DataFrame, n_edges_reconnect: int | None = None
 ) -> pd.Series:
+    # Import here since the import is very slow and expensive
+    from graspologic.partition import leiden
+
     component_edgelist = (
         component_edgelist.groupby(["upia", "upib"], observed=True)["count"]
         .count()