HNOCA fixes oregano

src/hnoca/mapping/mapper.py

@@ -33,8 +33,8 @@ def __init__(
             ref_model: The reference model to map the query dataset to.
         """
         # Check optional dependencies
-        check_deps("scvi-tools")
-        check_deps("scarches")
+        # check_deps("scvi-tools") # Comment as it is a function that fails to detect scvi-tools was installed
+        # check_deps("scarches") # Comment as it is a function that fails to detect scArches was installed
         # Import and store as attributes so other methods can use them
         import scarches
         import scvi  # local import assured by previous check

src/hnoca/mapping/wknn.py

@@ -53,13 +53,15 @@ def build_nn(  # noqa: D103
     ref,
     query=None,
     k=100,
-    use_rapids: bool = False,
+    weight: Literal["unweighted", "dist", "gaussian_kernel"] = "unweighted",
+    sigma=None,
+    use_rapids: bool = True, # Ensure that RAPIDS is used
 ):
     if query is None:
         query = ref
 
     if use_rapids:
-        check_deps("cuml")
+        # check_deps("cuml") # Comment check_deps() because the function is broken
         from cuml.neighbors import NearestNeighbors
 
         logger.info("Using cuML for neighborhood estimation on GPU.")
@@ -223,9 +225,9 @@ def estimate_presence_score(
         ref = ref_adata.obsm[use_rep_ref_trans_prop]
         ref_trans_prop = get_transition_prob_mat(ref, k=k_ref_trans_prop)
 
-    if split_by and split_by in query_adata.obs.columns:
+    if split_by in query_adata.obs.columns:
         presence_split = [
-            np.array(wknn[query_adata.obs[split_by] == x, :].sum(axis=0)).flatten()
+            np.array(wknn[query_adata.obs[split_by].to_numpy() == x, :].sum(axis=0)).flatten() # added to_numpy() for better compatibility
             for x in query_adata.obs[split_by].unique()
         ]
     else:
@@ -270,13 +272,14 @@ def estimate_presence_score(
     }
 
 
-def transfer_labels(ref_adata: sc.AnnData, query_adata: sc.AnnData, wknn, label_key: str = "celltype"):
+def transfer_labels(ref_adata: sc.AnnData, query_adata: sc.AnnData, wknn, label_key: str ="celltype"):
     """Transfer labels from reference to query data."""
     scores = pd.DataFrame(
         wknn @ pd.get_dummies(ref_adata.obs[label_key]),
         columns=pd.get_dummies(ref_adata.obs[label_key]).columns,
         index=query_adata.obs_names,
     )
+    scores["best_score"] = scores.max(1) # change order, first find the score then the label so no string is inputted among floats
     scores["best_label"] = scores.idxmax(1)
-    scores["best_score"] = scores.max(1)
+
     return scores