Lecture 6: Exploratory Data Analysis - A Case Study

You can do that with survey_df.loc[15:32]

How much data pruning to be carried out without affecting the Explanability of the ML results?

iam learning python too , let me know if you have good resources!!

my dataset has over 64000 rows and I need to drop the last 500 rows with elements from column 2

Actually, I am not sure of the advantages of using loc.

There should be a few ways to do this. The simplest I could work out now is
dataset["col_to_select"][:-500]

If you add a condition to the col_to_select as well, I have noticed that the indexing is not changed, so row 500:600 may not exist as a 100 rows anymore.

3 posts were split to a new topic: How to plot graphs & look for inconsistencies?

Ok. I will try that when the lecture is over. Thank you…

1 Like

Wan’t there a kaggle competion we should have entered as well besides the final course project? I guess it was removed, but does any one else remember this or am I finally going insane?

I was able to use jovianML library with WinPython and Pyside on Windows 7 64bit

Same here with windows 10 and standard python. I simply did pip install jovian (with admin privileges).

Did you manage to use “commit” without typing out the whole API key manually? Because the password cannot be pasted into the cmd line

Thank you for the course, it was really really interesting!! Well done!! Good job!!

Big thank you to @aakashns and Team!!

Thank you so much! Regards!

…and looking forward to the next one!!!

A big thank you to @aakashns and the whole course team. All the lectures in this course are awesome.

1 Like

A big THANK You to Aakash & JOVIAN team for such an amazing course. It’s well organized & well executed. Good amount of Knowledge shared… Once again thank you

od.download(‘countries-languages-spoken’) is giving this error,

ModuleNotFoundError: No module named ‘opendatasets.datasets.countries-languages-spoken’

Can anyone help?

You need to click the completed button on each of the weekly assignment page

I share these thoughts as well. I’m more confident I can do this work and am even more motivated to learn more. Thank you Aakash. BIG UPS to you and your team.

It’s working for me, but my concern is this dataset has 198 rows × 2 columns which becomes 1 column when I want to use country name as index. So someone can’t use this for the project?