I just download your code and
run examples/compare_all_models.py
but it has compile error as belows..
I think it may be a error(cuase I didn't change any thing from the code)?
And I wonder... I need to detect the percent of outliers. But the code has default contamination_percent. Could you remove it in your future version?
I think using the outlier score or probability... may support the detecting the default(?) contamination_percent using the train_set? of the input data in your future version of pyod. thanks.
12 fitting Locally Selective Combination (LSCP)
Traceback (most recent call last):
File "C:/Users/owner/Downloads/pyod-master/pyod-master/examples/compare_all_models.py", line 140, in
scores_pred = clf.decision_function(X) * -1
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\pyod\models\lscp.py", line 212, in decision_function
decision_scores = self._get_decision_scores(X)
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\pyod\models\lscp.py", line 273, in _get_decision_scores
local_pseudo_ground_truth, local_train_scores[:, d])[0]
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\scipy\stats\stats.py", line 3501, in pearsonr
raise ValueError('x and y must have length at least 2.')
ValueError: x and y must have length at least 2.
NeighborhoodCoding
commented
4 years ago
I just download your code and run examples/compare_all_models.py
but it has compile error as belows.. I think it may be a error(cuase I didn't change any thing from the code)? And I wonder... I need to detect the percent of outliers. But the code has default contamination_percent. Could you remove it in your future version? I think using the outlier score or probability... may support the detecting the default(?) contamination_percent using the train_set? of the input data in your future version of pyod. thanks.
12 fitting Locally Selective Combination (LSCP) Traceback (most recent call last): File "C:/Users/owner/Downloads/pyod-master/pyod-master/examples/compare_all_models.py", line 140, in
scores_pred = clf.decision_function(X) * -1
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\pyod\models\lscp.py", line 212, in decision_function
decision_scores = self._get_decision_scores(X)
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\pyod\models\lscp.py", line 273, in _get_decision_scores
local_pseudo_ground_truth, local_train_scores[:, d])[0]
File "C:\Users\owner\AppData\Local\Programs\Python\Python36\lib\site-packages\scipy\stats\stats.py", line 3501, in pearsonr
raise ValueError('x and y must have length at least 2.')
ValueError: x and y must have length at least 2.
Process finished with exit code 1