diff --git a/python/SphereNearestNeighbor.py b/python/SphereNearestNeighbor.py
deleted file mode 100644
index 2b892b1..0000000
--- a/python/SphereNearestNeighbor.py
+++ /dev/null
@@ -1,82 +0,0 @@
-from math import pi, sin, cos, acos
-from dataclasses import dataclass
-import numpy as np
-from sklearn.neighbors import BallTree, KDTree
-import sofar as sf
-import pyfar as pf
-from pathlib import Path
-
-sofa_path = Path(__file__).parent.parent / "assets"
-
-
-@dataclass
-class Point:
-    az: float
-    elev: float
-
-    def __init__(self, az, elev) -> None:
-        self.az = az / 360 * 2 * pi
-        self.elev = elev / 360 * 2 * pi
-
-    def dist(self, other):
-        d_lambda = other.az - self.az
-        inner = sin(self.elev) * sin(other.elev) + cos(self.elev) * cos(
-            other.elev
-        ) * cos(d_lambda)
-        d_rho = acos(inner)
-        return d_rho
-
-    def as_tuple(self, rev=False):
-        if rev:
-            return (self.elev, self.az)
-        else:
-            return (self.az, self.elev)
-
-    def __str__(self) -> str:
-        return f"Point(az={round(self.az, 2)}, elev={round(self.elev, 2)})"
-
-
-def construct_kdtree(points: list[Point], depth: int):
-    axis = depth % 3
-
-
-def find_nearest_naive(point: Point, points: list[Point]):
-    distances = [point.dist(p) for p in points]
-    idx = np.argmin(distances)
-    return distances[idx], idx
-
-
-sofa = sf.read_sofa(sofa_path / "pp1_HRIRs_measured.sofa")
-
-coords = pf.samplings.sph_lebedev(sh_order=10)
-
-to_find = pf.Coordinates(1, 0, 0)
-
-index, distance = coords.find_nearest(to_find)
-
-points = [Point(x[0], x[1]) for x in sofa.SourcePosition]
-# points = [Point(0, 0), Point(90, 0), Point(180, 0), Point(270, 0), Point(0, 90)]
-
-bt = BallTree([p.as_tuple(rev=True) for p in points], metric="haversine")
-
-testpoints = [
-    Point(x[0] * 360, x[1] * 180 - 90) for x in np.random.random_sample((200, 2))
-]
-
-points_successful = 0
-points_failed = 0
-for test_point in testpoints:
-    d_test, idx_test = bt.query([test_point.as_tuple(rev=True)])
-    d_correct, idx_correct = find_nearest_naive(test_point, points)
-    if idx_test[0][0] != idx_correct:
-        print(
-            f"failed for point {test_point}, idx_correct {idx_correct}, idx_test {idx_test[0][0]}, d_test {round(d_test[0], 2)}, d_correct {round(d_correct, 2)}"
-        )
-        points_failed += 1
-    else:
-        print(
-            f"success for point {test_point}, idx {idx_correct}, d_test {round(d_test[0][0], 2)}, d_correct {round(d_correct, 2)}"
-        )
-        points_successful += 1
-
-print(f"\n successful: {points_successful}, failed: {points_failed}")
diff --git a/python/TimeAlignment.ipynb b/python/TimeAlignment_first_try.ipynb
similarity index 100%
rename from python/TimeAlignment.ipynb
rename to python/TimeAlignment_first_try.ipynb