add cellcycle options

paper/ARI.py

@@ -0,0 +1,22 @@
+
+import pandas as pd
+import numpy as np
+
+result = pd.read_csv('/Volumes/salomonis2/LabFiles/Frank-Li/scTriangulate/pbmc10k/scTriangulate_present/user_choice_obs.txt',
+                     sep='\t',index_col=0)
+
+from sklearn.preprocessing import LabelEncoder
+from sklearn.metrics import adjusted_rand_score, adjusted_mutual_info_score, homogeneity_completeness_v_measure, v_measure_score
+from sklearn.metrics import normalized_mutual_info_score
+azimuth = LabelEncoder().fit_transform(result['azimuth'].values)
+leiden1 = LabelEncoder().fit_transform(result['leiden1'].values)
+leiden2 = LabelEncoder().fit_transform(result['leiden2'].values)
+leiden3 = LabelEncoder().fit_transform(result['leiden3'].values)
+final = LabelEncoder().fit_transform(result['reassign_prefix'].values)
+user = LabelEncoder().fit_transform(result['choice'].values)
+
+for test in [leiden1,leiden2,leiden3,final,user,azimuth]:
+    print('ARI:{}'.format(adjusted_rand_score(azimuth,test)))
+    print('AMI:{}'.format(adjusted_mutual_info_score(azimuth, test)))
+    print('V:{}'.format(v_measure_score(azimuth, test)))
+    print('NMI:{}'.format(normalized_mutual_info_score(azimuth,test)))