You are on page 1of 3

Implement simple vector addition in Tensorflow

AIM :
To write a python program to implement a simple vector addition in
tensorflow.

Algorithm:

1. Import TensorFlow: The program starts by importing the TensorFlow library


(import tensorflow as tf), which is a popular open-source machine learning
framework.
2. Define the Vector Addition Function: The vector_addition function takes two
input vectors ( a and b) as parameters.
3. Convert Input Lists to TensorFlow Tensors: Inside the function, the input lists ( a
and b) are converted into TensorFlow constant tensors using tf.constant().
4. Perform Vector Addition: The tf.add() function is used to perform element-wise
addition of the two tensors ( tensor_a and tensor_b), resulting in a new TensorFlow
tensor (result).
5. Return the Result Tensor: The function returns the TensorFlow tensor representing
the result of the vector addition.
6. Main Program Execution: The if __name__ == "__main__": block ensures that
the following code is executed only if the script is run directly, not if it's imported as
a module.
7. Define Sample Input Vectors: Two sample input vectors ( vector1 and vector2) are
defined as Python lists.
8. Perform Vector Addition: The vector_addition function is called with the sample
input vectors, and the result is stored in the variable result_vector.
9. Convert Result Tensor to NumPy Array: The result tensor ( result_vector) is
converted to a NumPy array using the .numpy() method.
10. Print the Results: The program prints the original vectors ( vector1 and vector2)
and the result of the vector addition ( result).
Program :
import tensorflow as tf

def vector_addition(a, b):


# Convert the input lists to TensorFlow tensors
tensor_a = tf.constant(a)
tensor_b = tf.constant(b)

# Perform the vector addition


result = tf.add(tensor_a, tensor_b)

return result

if __name__ == "__main__":
# Sample input vectors
vector1 = [1, 2, 3]
vector2 = [4, 5, 6]

# Perform vector addition using the function


result_vector = vector_addition(vector1, vector2)

# Open a TensorFlow session to execute the computation (not required in TF


2.x)
# with tf.compat.v1.Session() as sess:
# result = sess.run(result_vector)
# Instead, in TF 2.x, we can directly access the result as a NumPy array
result = result_vector.numpy()

print("Vector 1:", vector1)


print("Vector 2:", vector2)
print("Vector Addition Result:", result)

output:
Vector 1: [1, 2, 3]
Vector 2: [4, 5, 6]
Vector Addition Result: [5 7 9]

Result:
Thus the implementation of simple vector addition in tensorflow has
been executed successfully.

You might also like