Update logical issue in decision_tree.py (TheAlgorithms#13303)

Co-authored-by: Maxim Smolskiy <mithridatus@mail.ru>

Author: HarshPathak310

machine_learning/decision_tree.py

@@ -146,14 +146,13 @@ def predict(self, x):
  """
  if self.prediction is not None:
  return self.prediction
- elif self.left or self.right is not None:
+ elif self.left is not None and self.right is not None:
  if x >= self.decision_boundary:
  return self.right.predict(x)
  else:
  return self.left.predict(x)
  else:
- print("Error: Decision tree not yet trained")
- return None
+ raise ValueError("Decision tree not yet trained")
 
 
 class TestDecisionTree:
@@ -201,4 +200,4 @@ def main():
  main()
  import doctest
 
- doctest.testmod(name="mean_squarred_error", verbose=True)
+ doctest.testmod(name="mean_squared_error", verbose=True)