Lesson 1 - Linear Regression with Scikit Learn

Information on assignments and how to submit them will be shared in later lectures.

Here’s a link to the CSV file:
https://raw.githubusercontent.com/JovianML/opendatasets/master/data/medical-charges.csv

2 Likes

Guessing it is a library version issue, see if this fixes the issue, Plotly plots are blank on mybinder startup? - #5 by drscotthawley - repo help - Jupyter Community Forum

You can also run the code in google colab instead of binder, probably the graphs will be visible on google colab.

1 Like

Try running the cell, that should show up the graphs.

In the repository, it is named insurance.csv

2 posts were split to a new topic: What is the importance of R^2 value? Do we need to remove outliners to get good correlation?

for those students who are fresher and they have not so hand in python so if they complete the course with little late. is that ok with the certification and the course.

1 Like

@lavnishraghav00 First you will learn this course and after learn machine learning then your concept will be very clear

Yes, it is completely fine, you can first try out the other course as mentioned by @i16ph021.

Hi all, after completing the assignment i.e the notebook on linear regression and committing it to our profile. Where should we submit our assignment?

@i16ph021 if someone is already familiar with Pandas, does he still has to complete Zero to Pandas course?

@artaghal It’s very good. No, it is not necessary but the course will help you. To showcase certificate and course project.

2 Likes

Hey @haiderratlamwala786, welcome to the community.
The assignments for this course are not yet published and detail about the assignments/assignment submissions will be conveyed in later lectures.

Hey @artaghal, welcome to the community,
No you are not required to go through Zero To Pandas course to complete this course, but I would recommend you to go through the Pandas course too, cause it’s very interesting.
Happy Learning

1 Like

Thanks @birajde… surely i will be considering as per need…

1 Like

Hi, I have executed the code and I am getting the value of test loss to be less than value of training loss . Is this fine? What does this imply ?

Hey @muskan-gupta, welcome to the forum,
If the test loss is less than the training loss it is generally not a problem, I think the model can be trained more when the test loss is less, you can try to train the model more, but even without training the model if the results given by the model are good, you are good to go.
PS: Be sure not to include the target column(s), in the input columns!

1 Like

Thanks @birajde sir for clearing.

1 Like

Boston housing data should not be encouraged blindly, as it contains the racial profiling field. The dataset includes a Bk variable, a population of black.

2 posts were split to a new topic: What is the significance of ! and % prefix in notebook cells?