diff --git a/calculations/PMM.py b/calculations/PMM.py
index 8fd81ec..5553b4b 100644
--- a/calculations/PMM.py
+++ b/calculations/PMM.py
@@ -10,6 +10,7 @@ df['E'] = df['re_E'] + 1j * df['im_E']
 df['k'] = np.sqrt(df['E'])
 
 c0 = df[df['E'] == 0]['c'].values[0]
+df = df[df['c'] != c0] # remove the threshold point
 df['c'] = df['c'] - c0 # shift c to set c=0 at the exceptional point
 
 train_data = df[df['re_E'] < 0]
@@ -41,7 +42,7 @@ enforce_ep()
 
 # generate a set of c values to follow by subdividing the training cs
 subdivisions = 2
-c_steps = np.concatenate([np.linspace(start, stop, subdivisions, endpoint=False) for (start, stop) in zip(train_cs, train_cs[1:])])
+c_steps = np.concatenate([np.linspace(start, stop, subdivisions, endpoint=False) for (start, stop) in zip(np.insert(train_cs, 0, 0), train_cs)])
 c_steps = np.append(c_steps, train_cs[-1])
 c_steps = np.delete(c_steps, 0) # remove the first point (c=0)