# Machine Learning : Linear Regression using TensorFlow Python

Design, Develop and Train the model

In this course, we provide the step-by-step approach for building a Linear Regression model using TensorFlow with Python. In the beginning, we give a high-level introduction to Artificial Intelligence and Machine Learning. We develop the entire system in Google Colaboratory using TensorFlow. So, we have a lecture each on Introduction to Google Colaboratory and Introduction to TensorFlow. We develop the model to predict the price of the house from the size. We have the data for 100 houses with two attributes, house size, and house price. We first teach Python code to create the data, load it and check if the data are correctly loaded. We divide the data into Training and Testing data at a ratio of 80:20. We also introduce the importance of Data Normalization. After normalizing the data, we begin the process of building the model. We use the TensorFlow Gradient Descent method and train the model. We select the number of iterations to make the training error and testing error significantly low. After training the model we use the model for a new set of data. That is, we find the price of a new house whose size is given. We then extend the program for a problem with multiple variables. In this problem, we predict the price of the house from three attributes, plinth area, land area, and furnish-area. In the last lecture, elaborate more on training and test data and compute the same.

What you’ll learn

• Machine Learning – Linear Regression in TensorFlow with Python.
• TensorFlow model for Linear Regression.

Course Content

• Introduction to Artificial Intelligence and Machine Learning –> 2 lectures • 17min.
• The Building Blocks for Developing Program –> 2 lectures • 21min.
• Linear Regression Model –> 1 lecture • 13min.
• Data Preparation and Normalization –> 4 lectures • 38min.
• Linear Regression –> 3 lectures • 31min.
• Managing Data Files and of Pandas Dataframes –> 3 lectures • 32min.
• Linear Regression Model with More variables –> 2 lectures • 25min.
• Training and Testing Error in Machinle Learning Models –> 1 lecture • 14min. Requirements

• Programming Fundamentals.
• Be able to un derstand the code.

In this course, we provide the step-by-step approach for building a Linear Regression model using TensorFlow with Python. In the beginning, we give a high-level introduction to Artificial Intelligence and Machine Learning. We develop the entire system in Google Colaboratory using TensorFlow. So, we have a lecture each on Introduction to Google Colaboratory and Introduction to TensorFlow. We develop the model to predict the price of the house from the size. We have the data for 100 houses with two attributes, house size, and house price. We first teach Python code to create the data, load it and check if the data are correctly loaded. We divide the data into Training and Testing data at a ratio of 80:20. We also introduce the importance of Data Normalization. After normalizing the data, we begin the process of building the model. We use the TensorFlow Gradient Descent method and train the model. We select the number of iterations to make the training error and testing error significantly low. After training the model we use the model for a new set of data. That is, we find the price of a new house whose size is given. We then extend the program for a problem with multiple variables. In this problem, we predict the price of the house from three attributes, plinth area, land area, and furnish-area. In the last lecture, elaborate more on training and test data and compute the same.

Get Tutorial