need help .i did nt understand this gradient descent part need how to code and how to use in ML
I did nt understand this gradient descent part
hey @saieadara ,
firstly coming to gradient descent .
As a human , you learn things by experiencing them, initially you too perform poor on some things , but as you practice them again and again , you get to know how to work or use those things.
In short , you started learning about that particular thing.
Similar thing gradient descent , which helps our models or machines to understand the things .
initially the machine works very poor , but on some repeated iterations the machine starts learning some new concepts from the data provided. Gradient descent penalizes there learning by the amount of error they make , so larger the error your model make , the more it get penalized and more it learns.
I hope this helps you to get a bit of more understanding about gradient descent.
Now how to code is shown in detail in the videos , but to explain that a bit ,
we take a loss function that describe our learning , means for example we are have to do some task that depends upon some keypoints/variables , and there is a loss function ( we call it like that ) that tells how much accurately we have done this task.
So , initially we have a very bad performance , so we get the error as actual_value - our_calculated_value and as the loss function is equation depended on some variables ( on which we are working on ) , so to update our selves / simply penalizes ourselves , we take differentiation of this loss function with respect to the variables and then put the value of those variables we used currently , in this way after putting the values we get the penalization value to be done.
and this process gets repeated a number of times until or unless we are closest to the actual value .
I hope this helped you
Thank You and Happy Learning 
.
I hope I’ve cleared your doubt. I ask you to please rate your experience here
Your feedback is very important. It helps us improve our platform and hence provide you
the learning experience you deserve.
On the off chance, you still have some questions or not find the answers satisfactory, you may reopen
the doubt.