## Tensorflow Basics Tutorial Series – (Pt.1) Linear Regression

In this series we will learn the basics of tensorflow, starting from the beginning!

In this video, we’ll look at doing Linear Regression with Tensorflow

Reference repo: https://github.com/aymericdamien/TensorFlow-Examples
Github code: https://github.com/markjay4k/Tensorflow-Basics-Series/blob/master/Part%201%20-%20Linear%20Regression.ipynb
install Tensorflow-GPU
Discord Server: https://discordapp.com/invite/Zf8CSB3

Source

### Comment List

• Mark Jay
December 16, 2020

Hi Mark, great demo, very smooth with your Python. Question: If you have to know the model a-priori, why use tf? Why not just fit a regular linear model using OLS? Is the purpose to show gd in action using tf? How does one go about plotting the model solved by gd if they don't know it a-priori like you did here? Thanks

• Mark Jay
December 16, 2020

I think you need to use the dot product for the matrix multiplication W*X+B. tf.math.multiply is an element wise multiplication which is not the same. One yields a scalar and the other yields a matrix and will not work.

• Mark Jay
December 16, 2020

What version of tf did you use in this tutorial?

• Mark Jay
December 16, 2020

I'm trying to learn AI but didn't understand much of this tutorial ;-;

• Mark Jay
December 16, 2020

cool tutorial. thx so much!

• Mark Jay
December 16, 2020

A year ago! this is the best we can do is tutorials from a year ago!?!

• Mark Jay
December 16, 2020

epoch job!!

• Mark Jay
December 16, 2020

This tutorial is really epoch

• Mark Jay
December 16, 2020

This tutorial is really really good

• Mark Jay
December 16, 2020

this is an epoch tutorial

• Mark Jay
December 16, 2020

may I know the IDE you are using?

• Mark Jay
December 16, 2020

Good Job Mark, appreciate your effort

• Mark Jay
December 16, 2020

Great tutorial, this is what I needed
Could you explain why we need init = tf.global_variables_initializer( )?
In my understanding we've already initialised all the variables we need
or which global variables are we talking about?
Also why are we adding feed_dict to sess.run(optimizer, …) and sess.run(cost, …)
What is the purpose of feed_dict?
How do I choose the variables in feed_dict?

• Mark Jay
December 16, 2020

This series is really great,nice tutorial mate.

• Mark Jay
December 16, 2020

• Mark Jay
December 16, 2020

series videos are nice!!

• Mark Jay
December 16, 2020

• Mark Jay
December 16, 2020

You are a great instructor. Thank you very much.

• Mark Jay
December 16, 2020

Why tf.reduce_sum() is used in calculating cost? Without using it we can achieve the same. I have a doubt, MSE is calculated as "1/N * (∑(yi−(mxi+b))**2)" but, here we are using a single predicted value for calculating the cost each time when the optimizer is run instead of summation of all samples.

• Mark Jay
December 16, 2020

These Tutorials are great!! Love it!

• Mark Jay
December 16, 2020

Ya awesome, the best

• Mark Jay
December 16, 2020

• Mark Jay
December 16, 2020

I remember I tried setting up Tensorflow and CUDA back in February but it simply wouldn't work. Now you and Sentdex have fresh tutorials so I hope I'll be able to make it (I'll try on the summer vacation)

• Mark Jay
December 16, 2020

Hi, what is the IDE you are using?

• Mark Jay
December 16, 2020

Note to self.. "An epoch is one complete presentation of the data set to be learned."

• Mark Jay
December 16, 2020

Please keep this series going, there is not a single good tf tutorial on youtube. You can hit it off.

• Mark Jay
December 16, 2020

how will i change if i have multiple independent variable.. X consisting of age, price, date etc

• Mark Jay
December 16, 2020

Cool I’d also like to see some fortnite as well? What do you think?

• Mark Jay
December 16, 2020

Good tutorial, thanks.

• Mark Jay
December 16, 2020

Thank you 🙂 very nice tutorial!

• Mark Jay
December 16, 2020

Mr Jay, could you do a video using a martingale random walk scenario using tensorflow? I am new to python and i am looking to learn more…thanks this is an example of it by sentdex https://www.youtube.com/watch?v=ApGBEOK-jng