Machine Learning & Data Science Project - 3 : Feature Engineering (Real Estate Price Prediction)

1) Python
2) Numpy and Pandas for data cleaning
3) Matplotlib for data visualization
4) Sklearn for model building
5) Jupyter notebook, visual studio code and pycharm as IDE
6) Python flask for http server
7) HTML/CSS/Javascript for UI

Next video:

Popular Playlist:

[Picture_perfect_]

I don't know how people uses dislike button, for such a amazing content.
thanks for providing such a beautiful content and great explanation.

[CannonballCircuit]

Since the instructor is too busy teaching us valuable content, I thought I'd highlight the fact that he has a patreon everyone! The kind of content he puts out is totally worth $10 a month, let alone $1 a month, so to everyone who benefits from this: let's please contribute!

[flamboyantperson]

Feature Engineering is very important and you have explained it in a very nice and clear way.

[basotra]

Man, you are a genius, I love the way you explain everything. Keep bringing us these kinds of videos

[jepsmyt]

Nicely explained.

for unique value and their respective count in a data column we can use value_counts() istead of using unique/nunique/functions. let me know your comment on this.

[techtutsindia]

This video is helpful to learn new functions along with feature engineering.
To clean data and add new columns in the DataFrame.

[brijesh]

We can replace

by
df5.location.value_counts()
They will work same.
:)

[vishalrai]

sir, you are genius your concepts are crystal clear
I bet no coaching institute can teach better than you all are just money eaters

[subhabratanath]

Exploring new video is always fun, and the reason also succeed here, i learn some new things.

[pradyumnjoshi]

I am learning so many new things from your videos. Great work sir!!
Keep making those!

[galymzhankenesbekov]

very good, I am advising to all of my friends your videos. just as a suggestion please try to do more of these projects. I am interested also in soccer dataset, cars dataset. thanks a lot

[yingc]

Thank you so much for this sir. Have a wonderful day.

[fnflores]

All your tutorials are great! Thanks for sharing!

[adedejiadejumo]

Thank you for providing this tutorial, it is very useful. I have been following your tutorials for machine learning. In this particular video, I discovered that you performed feature engineering using the target variable. I am currently working on a project, and I performed feature engineering, After training the model and saved it as a Pickle file, I got a new dataset that is new to the model for validation purpose. Performing validation with the new dataset is difficult, if not impossible because the new dataset does not have a target variable for feature engineering, and the Trained model requires the same number of columns used for training. Please respond to this comment, thank you.

[bhavyaparikh]

You can use value_counts() for loxation column

[saksham_]

Sir, i have a query.
By using the target variable to engineer a new feature, aren't you "leaking" some of the data? I maybe wrong. Please correct me if so.

[mahindrakarabhijit]

I am getting error "unhashable type: 'list'" after running below code

df5.location = df5.location.apply(lambda x: 'others' if x in location_stats_less_than_10 else x)
len(df5.location.unique())

Kindly help

[adharapuramnavaneeth]

sir i guess we can directly use value_counts on location right? 4:52

[zishanafzal]

I have a question.
Len function is for calculating the length of the string. But how come here u are getting the count of occurance.
And secondly, what is token in the function that u defined.
Is it a keyword or a variable?

[utkarshkumar]

In 7:31
x: 'other' if x in loc_stats_less_than_10 else x)
It is giving error: argument of type int is not iterable
Please help me as i m a beginner....