update of univariate

python/src/univariate.py

@@ -40,23 +40,29 @@ class UNIVARIATE :
             self.covariates = scale(self.covariates) 
 
         self.use_resid = use_resid
-        if use_resid and isinstance(covariates, np.ndarray):
+
+        if use_resid and isinstance(self.covariates, np.ndarray):
             print("computing residuals")
-            self.outputs = np.apply_along_axis(lambda x : pt.adjust_covariates(x,covariates), axis = 0, arr = self.outputs) 
+            self.outputs = np.apply_along_axis(lambda x : pt.adjust_covariates(x,self.covariates), axis = 0, arr = self.outputs) 
 
-        self.p = None
+        self.p = []
         self.beta = None
 
     def test(self, **kwargs):
-        nan_mask = np.isnan(self.predictors).flatten()
+        self.p = []
+        for i in range(self.predictors.shape[1]):
+            print(i)
+            pred = self.predictors[:,i].reshape(-1,1)
+            nan_mask = np.isnan(pred).flatten()
             if ((self.covariates is not None) and (self.use_resid == False)) :
                 print(self.outputs.shape)
-            beta = linear_regression(self.outputs[~nan_mask], self.predictors[~nan_mask], self.covariates[~nan_mask])
+                beta = linear_regression(self.outputs[~nan_mask], pred[~nan_mask], self.covariates[~nan_mask])
             else :
-            beta = linear_regression(self.outputs[~nan_mask], self.predictors[~nan_mask])
+                beta = linear_regression(self.outputs[~nan_mask], pred[~nan_mask])
             self.beta = beta
-        res = ttest(beta, self.predictors[~nan_mask], self.outputs[~nan_mask])
-        self.p = res
+            res = ttest(beta, pred[~nan_mask], self.outputs[~nan_mask])
+            self.p += [res]
+
 
     def plot(self, show = True):
         fig = make_subplots(rows = 1,cols=1)