Scribd
Upload
9 views

Tensor Flow

Uploaded by

anushkarokade21
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views

Tensor Flow

Uploaded by

anushkarokade21
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

TensorFlow

TensorFlow is a library that helps engineers build


and train deep learning models.
We can use TensorFlow to train simple to complex neural
networks using large sets of data.
TensorFlow enables us to quickly and easily build powerful AI
models with high accuracy and performance.
What is a Tensor?
A simple explanation would be that a tensor is a multi-
dimensional array.

Scalar, Vector, Matrix and Tensor

A scalar is a single number. A vector is an array of numbers. A


matrix is a 2-dimensional array. A tensor is an n-dimensional
array.
In TensorFlow, everything can be considered a tensor including
a scalar. A scalar would be a tensor of dimension 0, a vector of
dimension 1, and a matrix of dimension 2.
What is TensorFlow?
TensorFlow is an open-source software library for building
neural networks. Google Brain team was the one who built it
and it is the most popular deep learning library in the market
today.

You can use TensorFlow to build AI models including image


and speech recognition, natural language processing, and
predictive modeling.

TensorFlow and Keras

TensorFlow has a high-level API called Keras. Keras was a


standalone project which is now available within the TensorFlow
library. Keras makes it easy to define and train models while
TensorFlow provides more control over the computation.
TensorFlow supports a wide range of hardware, including
CPUs, GPUs, and TPUs. TPUs are Tensor processing Unites,
built specifically to work with Tensors and TensorFlow.
We can also run TensorFlow on mobile devices and IoT devices
using TensorFlow Lite.

How to Build Tensors with TensorFlow

Importing TensorFlow and printing out the version.

import tensorflow as tf
print(tf.__version__)

OUTPUT:
2.9.2
First create a scalar using tf.constant.
scalar = tf.constant(7)
print(scalar)
print(scalar.ndim)

OUTPUT:
tf.Tensor(7, shape=(), dtype=int32)
0

Create a vector
vector = tf.constant([10,10])
print(vector)
print(vector.ndim)

OUTPUT:
tf.Tensor([10 10], shape=(2,), dtype=int32)
1

Creating a Matrix
matrix = tf.constant([
[10,11],
[12,13]
])
print(matrix)
print(matrix.ndim)

OUTPUT:
tf.Tensor(
[[10 11]
[12 13]], shape=(2, 2), dtype=int32)
2

Tensors have a default datatype of int32.


Simple calculations using the tensor

basic_tensor = tf.constant([[10,11],[12,13]])
print(basic_tensor)

OUTPUT:

tf.Tensor(
[[10 11]
[12 13]], shape=(2, 2), dtype=int32)

We can add, subtract, multiply, and divide every value in a


tensor using the basic operators.
print(basic_tensor + 10)
print(basic_tensor - 10)
print(basic_tensor * 10)
print(basic_tensor / 10)

OUTPUT:
tf.Tensor(
[[20 21]
[22 23]], shape=(2, 2), dtype=int32)
tf.Tensor(
[[0 1]
[2 3]], shape=(2, 2), dtype=int32)
tf.Tensor(
[[100 110]
[120 130]], shape=(2, 2), dtype=int32)
tf.Tensor(
[[1. 1.1]
[1.2 1.3]], shape=(2, 2), dtype=float64)
Matrix Multiplication
tensor_011 = tf.constant([[2,2],[4,4]])
tensor_012 = tf.constant([[2,3],[4,5]])

print(tf.matmul(tensor_011,tensor_012))

OUTPUT:
tf.Tensor(
[[12 16]
[24 32]], shape=(2, 2), dtype=int32)

To find the minimum and maximum values


tensor_013 = tf.constant([
[1,2,3],
[4,5,6],
[7,8,9]
],dtype='float32')
print(tf.reduce_min(tensor_013))
print(tf.reduce_max(tensor_013))
print(tf.reduce_sum(tensor_013))

OUTPUT:
tf.Tensor(1.0, shape=(), dtype=float32)
tf.Tensor(9.0, shape=(), dtype=float32)
tf.Tensor(45.0, shape=(), dtype=float32)

Find the square, square root, and log of each value in a tensor.
print(tf.sqrt(tensor_013))
print(tf.square(tensor_013))
print(tf.math.log(tensor_013))
OUTPUT:
tf.Tensor(
[[1. 1.4142135 1.7320508]
[2. 2.236068 2.4494898]
[2.6457512 2.828427 3. ]], shape=(3, 3), dtype=float32)
tf.Tensor(
[[ 1. 4. 9.]
[16. 25. 36.]
[49. 64. 81.]], shape=(3, 3), dtype=float32)
tf.Tensor(
[[0. 0.6931472 1.0986123]
[1.3862944 1.609438 1.7917595]
[1.9459102 2.0794415 2.1972246]], shape=(3, 3), dtype=float32)

Conclusion
Tensorflow is a powerful library to build deep-learning models. It
has all the tools we need to construct neural networks to solve
problems like image classification, sentiment analysis, stock
market predictions, etc.

You might also like