Update simple_adaline.py

neural_network/simple_adaline.py

@@ -1,81 +1,110 @@
-def wght_cng_or(wgt, T, al):
- O = wgt[0] * 0 + wgt[1] * 0
- if O <= T:
- Ol = wgt[0] * 0 + wgt[1] * 1
- if Ol >= T:
- Ole = wgt[0] * 1 + wgt[1] * 0
- if Ole >= T:
- Ola = wgt[0] * 1 + wgt[1] * 1
- if Ola >= T:
- return wgt
+
+from typing import List, Tuple
+
+def weight_change_or(weight: List[float], threshold: float, learning_rate: float) -> List[float]:
+ output = weight[0]*0 + weight[1]*0
+ if output <= threshold:
+ output_left = weight[0]*0 + weight[1]*1
+ if output_left >= threshold:
+ output_left_down = weight[0]*1 + weight[1]*0
+ if output_left_down >= threshold:
+ output_all = weight[0]*1 + weight[1]*1
+ if output_all >= threshold:
+ return weight
  else:
- wgt[0] = wgt[0] + al * 1 * 1
- wgt[1] = wgt[1] + al * 1 * 1
- return wght_cng_or(wgt, T, al)
+ weight[0] = weight[0] + learning_rate*1*1
+ weight[1] = weight[1] + learning_rate*1*1
+ return weight_change_or(weight, threshold, learning_rate)
  else:
- wgt[0] = wgt[0] + al * 1 * 1
- wgt[1] = wgt[1] + al * 1 * 0
- return wght_cng_or(wgt, T, al)
+ weight[0] = weight[0] + learning_rate*1*1
+ weight[1] = weight[1] + learning_rate*1*0
+ return weight_change_or(weight, threshold, learning_rate)
  else:
- wgt[0] = wgt[0] + al * 1 * 0
- wgt[1] = wgt[1] + al * 1 * 1
- return wght_cng_or(wgt, T, al)
+ weight[0] = weight[0] + learning_rate*1*0
+ weight[1] = weight[1] + learning_rate*1*1
+ return weight_change_or(weight, threshold, learning_rate)
  else:
- T += al
- return wght_cng_or(wgt, T, al)
+ threshold += learning_rate
+ return weight_change_or(weight, threshold, learning_rate)
 
 
-def wght_cng_and(wgt, T, al):
- O = wgt[0] * 0 + wgt[1] * 0
- if O <= T:
- Ol = wgt[0] * 0 + wgt[1] * 1
- if Ol <= T:
- Ole = wgt[0] * 1 + wgt[1] * 0
- if Ole <= T:
- Ola = wgt[0] * 1 + wgt[1] * 1
- if Ola >= T:
- return wgt
+def weight_change_and(weight: List[float], threshold: float, learning_rate: float) -> List[float]:
+ output = weight[0]*0 + weight[1]*0
+ if output <= threshold:
+ output_left = weight[0]*0 + weight[1]*1
+ if output_left <= threshold:
+ output_left_down = weight[0]*1 + weight[1]*0
+ if output_left_down <= threshold:
+ output_all = weight[0]*1 + weight[1]*1
+ if output_all >= threshold:
+ return weight
  else:
- wgt[0] = wgt[0] + (al * 1 * 1)
- wgt[1] = wgt[1] + (al * 1 * 1)
- return wght_cng_and(wgt, T, al)
+ weight[0] = weight[0] + (learning_rate*1*1)
+ weight[1] = weight[1] + (learning_rate*1*1)
+ return weight_change_and(weight, threshold, learning_rate)
  else:
- wgt[0] = wgt[0] - (al * 1 * 1)
- wgt[1] = wgt[1] - (al * 1 * 0)
- return wght_cng_and(wgt, T, al)
+ weight[0] = weight[0] - (learning_rate*1*1)
+ weight[1] = weight[1] - (learning_rate*1*0)
+ return weight_change_and(weight, threshold, learning_rate)
  else:
- wgt[0] = wgt[0] - (al * 1 * 0)
- wgt[1] = wgt[1] - (al * 1 * 1)
- return wght_cng_and(wgt, T, al)
+ weight[0] = weight[0] - (learning_rate*1*0)
+ weight[1] = weight[1] - (learning_rate*1*1)
+ return weight_change_and(weight, threshold, learning_rate)
  else:
- T += al
- return wght_cng_and(wgt, T, al)
+ threshold += learning_rate
+ return weight_change_and(weight, threshold, learning_rate)
+
 
+def and_gate(weight: List[float], input_a: int, input_b: int, threshold: float, learning_rate: float) -> int:
+ """
+ This function implements the AND gate using the Adaline algorithm.
 
-def and_gate(wgt, A, B, T, al):
- wgt = wght_cng_and(wgt, T, al)
- O = wgt[0] * A + wgt[1] * B
- if O >= T:
+ Args:
+ weight (List[float]): The weights for the Adaline algorithm.
+ input_a (int): The first input value.
+ input_b (int): The second input value.
+ threshold (float): The threshold value for the Adaline algorithm.
+ learning_rate (float): The learning rate for the Adaline algorithm.
+
+ Returns:
+ int: The output of the AND gate.
+ """
+ weight = weight_change_and(weight, threshold, learning_rate)
+ output = weight[0]*input_a + weight[1]*input_b
+ if output >= threshold:
  return 1
  else:
  return 0
 
 
-def or_gate(wgt, A, B, T, al):
- wgt = wght_cng_or(wgt, T, al)
- O = wgt[0] * A + wgt[1] * B
- if O >= T:
+def or_gate(weight: List[float], input_a: int, input_b: int, threshold: float, learning_rate: float) -> int:
+ """
+ This function implements the OR gate using the Adaline algorithm.
+
+ Args:
+ weight (List[float]): The weights for the Adaline algorithm.
+ input_a (int): The first input value.
+ input_b (int): The second input value.
+ threshold (float): The threshold value for the Adaline algorithm.
+ learning_rate (float): The learning rate for the Adaline algorithm.
+
+ Returns:
+ int: The output of the OR gate.
+ """
+ weight = weight_change_or(weight, threshold, learning_rate)
+ output = weight[0]*input_a + weight[1]*input_b
+ if output >= threshold:
  return 1
  else:
  return 0
 
 
 weight = [1.2, 0.6]
 weight2 = [1.2, 0.6]
-T = 1
-al = 0.5
-A, B = input("Input the value of A and B:").split()
-A = int(A)
-B = int(B)
-print("\nThe output of OR is:", or_gate(weight, A, B, T, al))
-print("\nThe output of AND is:", and_gate(weight2, A, B, T, al))
+threshold = 1
+learning_rate = 0.5
+input_a, input_b = input("Input the value of A and B:").split()
+input_a = int(input_a)
+input_b = int(input_b)
+print("\nThe output of OR is:", or_gate(weight, input_a, input_b, threshold, learning_rate))
+print("\nThe output of AND is:", and_gate(weight2, input_a, input_b, threshold, learning_rate))