Merge branch 'release/6.1.1'

Author: hbredin

doc/source/changelog.rst

@@ -2,10 +2,15 @@
 Changelog
 #########
 
+Version 6.1.1 (2025-12-07)
+~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- fix: fix warning about invalid espace sequence
+
 Version 6.1.0 (2025-09-19)
 ~~~~~~~~~~~~~~~~~~~~~~~~~~
 
--  BREAKING: update `STMLoader` to return `meeteval.io.SegLST` instances
+- BREAKING: update `STMLoader` to return `meeteval.io.SegLST` instances
 
 Version 6.0.0 (2025-09-09)
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

src/pyannote/database/loader.py

@@ -87,7 +87,7 @@ def load_trial(file_trial):
         List of trial
     """
 
-    trials = pd.read_table(file_trial, sep="\s+", names=["reference", "uri1", "uri2"])
+    trials = pd.read_table(file_trial, sep=r"\s+", names=["reference", "uri1", "uri2"])
 
     for _, reference, uri1, uri2 in trials.itertuples():
         yield {"reference": reference, "uri1": uri1, "uri2": uri2}
@@ -323,7 +323,7 @@ def __init__(self, mapping: Path):
         dtype = {
             "uri": str,
         }
-        self.data_ = pd.read_csv(mapping, names=names, dtype=dtype, sep="\s+")
+        self.data_ = pd.read_csv(mapping, names=names, dtype=dtype, sep=r"\s+")
 
         # get colum 'value' dtype, allowing us to acces it during subset
         self.dtype = self.data_.dtypes["value"]

src/pyannote/database/util.py

@@ -179,7 +179,7 @@ def load_rttm(file_rttm, keep_type="SPEAKER"):
         file_rttm,
         names=names,
         dtype=dtype,
-        sep="\s+",
+        sep=r"\s+",
         keep_default_na=True,
     )
 
@@ -213,7 +213,7 @@ def load_stm(file_stm):
     dtype = {"uri": str, "speaker": str, "start": float, "end": float}
     data = pd.read_csv(
         file_stm,
-        sep="\s+",
+        sep=r"\s+",
         usecols=[0, 2, 3, 4],
         dtype=dtype,
         names=list(dtype),
@@ -250,7 +250,7 @@ def load_mdtm(file_mdtm):
         file_mdtm,
         names=names,
         dtype=dtype,
-        sep="\s+",
+        sep=r"\s+",
         keep_default_na=False,
     )
 
@@ -281,7 +281,7 @@ def load_uem(file_uem):
 
     names = ["uri", "NA1", "start", "end"]
     dtype = {"uri": str, "start": float, "end": float}
-    data = pd.read_csv(file_uem, names=names, dtype=dtype, sep="\s+")
+    data = pd.read_csv(file_uem, names=names, dtype=dtype, sep=r"\s+")
 
     timelines = dict()
     for uri, parts in data.groupby("uri"):
@@ -306,7 +306,7 @@ def load_lab(path, uri: str = None) -> Annotation:
 
     names = ["start", "end", "label"]
     dtype = {"start": float, "end": float, "label": str}
-    data = pd.read_csv(path, names=names, dtype=dtype, sep="\s+")
+    data = pd.read_csv(path, names=names, dtype=dtype, sep=r"\s+")
 
     annotation = Annotation(uri=uri)
     for i, turn in data.iterrows():