Python script Table11

scripts/StepMix_Article_Table11.py

@@ -0,0 +1,164 @@
+import pandas as pd
+import numpy as np
+from stepmix.stepmix import StepMix
+import time
+
+#Carcinoma
+carci_DF = pd.read_csv("Data/carcinoma.csv")
+carci_MM = carci_DF.iloc[:, [1,2,3,4,5,6,7]]
+carci_MM -= 1 #to obtain binary indicators
+
+t0 = time.time()
+model_simple = StepMix(n_components=3, 
+                       measurement='binary', 
+                       random_state=123,  
+                       verbose=1)
+
+model_simple.fit(carci_MM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+#Simulated data: Distal outcome
+Datasim_Dist_DF = pd.read_csv("Data/datasim_Dist.csv")
+Datasim_Dist_MM = Datasim_Dist_DF.iloc[:, [1,2,3,4,5,6]]
+Datasim_Dist_SM = Datasim_Dist_DF.iloc[:, [7]]
+
+t0 = time.time()
+model_Dist_SM = StepMix(n_components=3, 
+                        measurement='binary',
+                        structural='gaussian_unit',
+                        random_state=123,
+                        n_steps=1,
+                        verbose=1)
+
+model_Dist_SM.fit(Datasim_Dist_MM, Datasim_Dist_SM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+
+#Simulated data: Covariate
+Datasim_cov_DF = pd.read_csv("Data/datasim_cov.csv")
+Datasim_cov_MM = Datasim_cov_DF.iloc[:, [1,2,3,4,5,6]]
+Datasim_cov_SM = Datasim_cov_DF.iloc[:, [7]]
+
+covariate_params = {
+    "method": "newton-raphson", 
+    "max_iter": 1,
+    "intercept": True,
+}
+
+t0 = time.time()
+model_cov_SM = StepMix(n_components=3, 
+                       measurement='binary',
+                       structural='covariate', 
+                       structural_params=covariate_params,
+                       random_state=123,
+                       n_steps=1,
+                       verbose=1)
+
+model_cov_SM.fit(Datasim_cov_MM, Datasim_cov_SM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+#IRIS
+Iris_DF = pd.read_csv("Data/iris.csv")
+Iris_MM = Iris_DF.iloc[:, [1,2,3,4]]
+
+
+t0 = time.time()
+model_iris_MM = StepMix(n_components=3, 
+                        measurement='continuous', 
+                        random_state=123,
+                        verbose=1)
+
+model_iris_MM.fit(Iris_MM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+#Diabetes
+Diabetes_DF = pd.read_csv("Data/diabetes.csv")
+Diabetes_MM = Diabetes_DF.iloc[:, [2,3,4]]
+
+Diabetes_SM = Diabetes_DF.iloc[:, [1]]
+SM_param_integer = {"Normal": 0, "Chemical": 1, "Overt": 2}
+Diabetes_SM['class'] = Diabetes_SM['class'].map(SM_param_integer)
+
+t0 = time.time()
+model_diabetes_SM = StepMix(n_components=3, 
+                            measurement='continuous', 
+                            structural='categorical',
+                            random_state=123,
+                            n_steps=1,
+                            verbose=1)
+
+model_diabetes_SM.fit(Diabetes_MM,Diabetes_SM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+#Banknote
+Banknote_DF = pd.read_csv("Data/banknote.csv")
+Banknote_MM = Banknote_DF.iloc[:, [2,3,4,5,6,7]]
+
+Banknote_SM = Banknote_DF.iloc[:, [1]]
+SM_param_integer2 = {"genuine": 0, "counterfeit": 1}
+Banknote_SM['Status'] = Banknote_SM['Status'].map(SM_param_integer2)
+
+covariate_params = {
+    "method": "newton-raphson", 
+    "max_iter": 1,
+    "intercept": True,
+}
+
+t0 = time.time()
+model_banknote_SM = StepMix(n_components=2, 
+                            measurement='continuous', 
+                            structural='covariate',
+                            structural_params=covariate_params,
+                            random_state=123,
+                            n_steps=1,
+                            verbose=1)
+
+model_banknote_SM.fit(Banknote_MM, Banknote_SM)
+t1 = time.time()
+
+total = t1-t0
+round(total, 5)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+