According to "TensorFlow is an end-to-end open source platform for machine learning. It has a comprehensive, flexible ecosystem of tools, libraries and community resources that lets researchers push the state-of-the-art in ML and developers easily build and deploy ML powered applications". Basically, it is library that makes it easier to build machine learning models. TensorFlow allows to easily create deep neural networks to use in computer vision, machine learning and Data mining.  The name TensorFlow derives from the operations that neural networks perform on multidimensional data arrays, which are referred to as tensors.

Brief explanation of What Machine Learning

Machine learning is a new programming paradigm in which you train the computers to perform actions. Machine learning consist of giving the computer label data and allow the computer to learn the patterns from the data.  So instead of us as developers figuring out the rules of problem, what we will do is we can get a bunch of examples for what we want to see and then have the computer figure out the rules. So, consider this example, fruit recognition. If I'm building a device that detects different fruits and i have label pictures data for various stage of ripeness for some fruit. A machine learning algorithm can figure out the specific patterns in each set of data that determines the distinctiveness of each. That's what's so powerful and exciting about this programming paradigm. Machine learning falls into three categories supervise learning and , deep learning, and reinforcement learning.

Supervisor learning

ML systems learn how to combine input to produce useful predictions on never-before-seen data. Supervised learning produces functions based on the labeled input data to learn. Supervised learning consists of the following


label is thing we are predicting. The label could be a type of shoes wheat, the kind of shoes is shown in a picture, the meaning of an audio clip, or about anything.


feature is an input variable. A simple machine learning project might use a single feature, while a more sophisticated machine learning project could use millions of features, specified as:

Label and Features Example

Example of dataset - I choice the column to brand to be my label that means we trying to predict the brand of car then the other columns become the features . These will be the input variables for my

price brand model year title_status mileage color vin lot state country condition
6300 toyota cruiser 2008 clean vehicle 274117 black   jtezu11f88k007763 159348797 new jersey  usa 10 days left
2899 ford se 2011 clean vehicle 190552 silver   2fmdk3gc4bbb02217 166951262 tennessee  usa 6 days left
5350 dodge mpv 2018 clean vehicle 39590 silver   3c4pdcgg5jt346413 167655728 georgia  usa 2 days left
25000 ford door 2014 clean vehicle 64146 blue   1ftfw1et4efc23745 167753855 virginia  usa 22 hours left
27700 chevrolet 1500 2018 clean vehicle 6654 red   3gcpcrec2jg473991 167763266 florida  usa 22 hours left
5700 dodge mpv 2018 clean vehicle 45561 white   2c4rdgeg9jr237989 167655771 texas  usa 2 days left
7300 chevrolet pk 2010 clean vehicle 149050 black   1gcsksea1az121133 167753872 georgia  usa 22 hours left

Unsupervised learning

 is a type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses. The most common unsupervised learning method is cluster analysis, which is used for exploratory data analysis to find hidden patterns or grouping in data.



A model is the function that is produce after applying machine learning algorithms to a dataset. this model is defining the relationship between features and label. For example, a spam detection model might associate certain features strongly with "spam". Let's highlight two phases of a model's life:

  • Training means creating or learning the model. That is, you show the model labeled examples and enable the model to gradually learn the relationships between features and label.
  • Inference means applying the trained model to unlabeled examples. That is, you use the trained model to make useful predictions (y'). For example, during inference, you can predict medianHouseValue for new unlabeled examples.

Regression vs. classification

Most supervising learning problems can be defined into two type problems regression and classification.

regression model predicts numeric values (continuous values). For example, regression models make predictions that answer questions like the following:

  • What is the value of a house in Jamaica?
  • What is the probability that a stock price will increase?

classification model predicts discrete values. For example, classification models make predictions that answer questions like the following:

  • Is a given person email message spam or not spam?
  • Is this an image of a melon, a apple, or a yam?

Deep learning

Deep learning is an artificial intelligence function that imitates the workings of the human brain in processing data and creating patterns for use in decision making. Deep learning is a subset of machine learning in artificial intelligence (AI) that has networks capable of learning unsupervised from data that is unstructured or unlabeled. Also known as deep neural learning or deep neural network.

Reinforcement Learning

Reinforcement Learning is defined as a Machine Learning method that is concerned with how software agents should take actions in an environment. Reinforcement Learning is a part of the deep learning method that helps you to maximize some portion of the cumulative reward.

Features of TensorFlow

TensorFlow APIs are arranged hierarchically, with the high-level APIs built on the low-level APIs. Machine learning researchers use the low-level APIs to create and explore new machine learning algorithms. In this class, you will use a high-level API named tf.keras to define and train machine learning models and to make predictions. tf.keras is the TensorFlow variant of the open-source Keras API.

Videos to use to do this introduction on tensorflow

Working through ‘Hello World’ in TensorFlow and Python

The ‘Hello World’ of neural networks


Read More