Closed
Description
Here is a quick example:
dist = 'seuclidean'
D = pairwise_distances(X[:400,:], metric = dist )
D = np.nan_to_num(D)
eps = np.percentile(D, 0.6)
print eps
dscan = DBSCAN(eps=eps, min_samples=5, metric = dist)
y = dscan.fit_predict(X)
File "cluster.py", line 91, in cluster
y = dscan.fit_predict(X)
File "/usr/local/lib/python2.7/dist-packages/sklearn/base.py", line 345, in fit_predict
self.fit(X)
File "/usr/local/lib/python2.7/dist-packages/sklearn/cluster/dbscan_.py", line 248, in fit
clust = dbscan(X, **self.get_params())
File "/usr/local/lib/python2.7/dist-packages/sklearn/cluster/dbscan_.py", line 101, in dbscan
neighbors_model.fit(X)
File "/usr/local/lib/python2.7/dist-packages/sklearn/neighbors/base.py", line 661, in fit
return self._fit(X)
File "/usr/local/lib/python2.7/dist-packages/sklearn/neighbors/base.py", line 234, in _fit
**self.effective_metric_params_)
File "binary_tree.pxi", line 1062, in sklearn.neighbors.ball_tree.BinaryTree.__init__ (sklearn/neighbors/ball_tree.c:8037)
File "dist_metrics.pyx", line 280, in sklearn.neighbors.dist_metrics.DistanceMetric.get_metric (sklearn/neighbors/dist_metrics.c:4066)
File "dist_metrics.pyx", line 435, in sklearn.neighbors.dist_metrics.SEuclideanDistance.__init__ (sklearn/neighbors/dist_metrics.c:5414)
TypeError: __init__() takes exactly 1 positional argument (0 given)
If I calculate distance matrix with it, the everything is just fine, but I want to use kd_tree or something to speed things up.