Data Visualization BATTLE!

Can I create 3 random data visualizations with python? Yeah. Are they any good? I'll let you decide.
Notebook with code: www.kaggle.com/robikscube/you...
Follow me on twitch for live coding streams: / medallionstallion_
My other videos:
Speed Up Your Pandas Code: • Make Your Pandas Code ...
Speed up Pandas Code: • Make Your Pandas Code ...
Intro to Pandas video: • A Gentle Introduction ...
Exploratory Data Analysis Video: • Exploratory Data Analy...
Working with Audio data in Python: • Audio Data Processing ...
Efficient Pandas Dataframes: • Speed Up Your Pandas D...
* KZbin: youtube.com/@robmulla?sub_con...
* Discord: / discord
* Twitch: / medallionstallion_
* Twitter: / rob_mulla
* Kaggle: www.kaggle.com/robikscube
== Music ==
Video Link: • 15 Minutes Royalty Fre...
Music provided by Johnusworld - Music for Video Creators
Creative Commons - Attribution 3.0 Unported
====================================
#python #datascience #datavisualization

[@DataCraftsman]

I really liked this. Making good graphs is hard work and this style of video shows a lot of tricks in a short amount of time. I learned at least 5 tricks from this video!

[@robmulla]

Glad you liked it! I was trying something different with this video and was worried people might not like the fast paced style. Happy you learned something new.

[@simoneparvizi775]

Please more Data visualization tutorials, also with only categorical variables. There is literally NOTHING on youtube and I love your format and your explanations. Keep up the great work you'll definitely hit 1M

[@nirbhay_raghav]

This is just mindblowing. Please keep them coming. I suck at making plots. Specially with all the other things to handle in them. Looking at his video gives me confidence that it is not THAT hard but it can be done if you practice it daily. And after some point it becomes instinct.

[@robmulla]

I apprecaite that Nirbhay! I thought this video was fun to make. I didn't know how it would turn out before I started but fun to hear you found it encouraging. Learning a little every day is key!

[@user-kr4fe1gg7r]

Thank you so much! I’m new into data science and I’ve been looking for videos that show raw process of data science and failed. Your videos are so good in that they show how a data professional actually do to solve these problems!

[@robmulla]

Love that you find my videos helpful. There are infinite ways to approach these types of problems but at least I can show how I do it!

[@leibaleibovich5806]

Thank you for the video and livestreams! The energy is absolutely amazing! In my opinion, you are literally doing some magic on air. While watching a livestream, I must set everything else aside, because it is difficult to get un-glued during the stream! I would like to ask you few questions, if I may. As someone, who contemplates a career change, I would like to understand what I need to study. 1) First and foremost, what libraries do you use the most (say, 80% of your time)? I would assume pandas, numpy, matplotlib, seaborn, sci-kit learn. What else have I missed? 2) Without going into specifics, on your daily job, what is the end-product of your data analysis? Is it an intelligence brief, a memo, a report? I come from a research background, so I my case an end product of all experiments is a paper in a peer-review journal. Let's say, I want to do a practice project for my portfolio and I want to make it look like a finished product, not just handful of charts. How would I do it? 3) Have you ever been on hiring committees? I am wondering, what you will be looking for in a candidate as the data analyst? In my case, I do not have a Computer Sci. degree or Data Sci. degree, how can I make myself competitive in this market? Many thanks for your answers!

[@robmulla]

Hey Leiba. I really appreciate your comments and feedback. Happy to hear you enjoy watching the live streams. Great questions. I can't answer them in full here, maybe at some point I'll do an AMA on stream, but for how I can say: 1) You got most of them, depends if I'm modeling but: pandas/numpy/matplotlib/seaborn/lightgbm/xgboost/pytorch/tensorflow. 2) It depends, sometimes a product and some times analytics briefs, most of the time things are presented at some point in a powerpoint deck. 3) Yes, I'd say the biggest thing to getting a foot in the door is knowing someone. People are more likely to hire someone who is endorsed by a person they trust. After that I'd say do something to stand out like win a kaggle competition or submit to a journal or write a data science blog. Hope that helps!

[@williamgriffin61]

thanks for the great video Rob! so helpful for our work.

[@robertmagyar621]

We're enjoying it Csanád, you are the best

[@robmulla]

Glad you enjoy it!

[@olgierd245]

Pretty cool. I liked that. I'm surely going to dive more into your videos

[@robmulla]

Please do! Thanks for the feedback.

[@thomasrademaker94]

Good stuff, very realistic experience. It is clear that even for an expert as yourself data wrangling takes time and effort.

[@robmulla]

So true. Thanks for watching.

[@gabefosse2050]

great video, keep it up. would like to hear more about your thought process and problem definitions

[@robmulla]

Thanks Gabe! Will do! I tried to make this video more fast paced, while my live streams are usually slowly working through a problem. Hope you subscribed so you can check one of those out sometime.

[@darren7193]

Excellent vid plz make another one

[@srujithreddy1643]

This battle is awesome

[@robmulla]

Thanks!

[@Levy957]

That's soo good!! Keep up!

[@robmulla]

Apprecaite the kind words Levy.

[@zheshipeng]

Entertaining and informative

[@jrdz12]

This is a great format. I enjoyed the video. Where do the datasets come from? I deal with a lot of timeseries data (data acquisition systems gathering second by second data)...

[@robmulla]

Glad you asked. I have the notebook linked in the description. The datasets are all ones I made on my live streams and are linked in that notebook.

[@jrdz12]

@@robmulla that's great. Thanks!!

[@nickolastradess]

Please do more of these videos!

[@robmulla]

Thanks Nick! I wasn't sure people would like this one but glad to hear you did.

[@ratulr]

Loved this video

[@robmulla]

So glad you did. Might try to make some more like it.

[@ohmygoddiosmio9138]

impressive knowledgeof the subject !

[@robmulla]

Thanks. Hope you learned something!

[@AndreWicaksana-mm8rb]

I still wondering why do you use max - min for the first data set ?

[@jonahjohnbaba]

Wow this is going to be fun

[@robmulla]

Hope you enjoyed it. A little different than my other videos so far.

[@sch0ll1]

Thank's for the video :) But one question, why do you start your path string with '..'?

[@robmulla]

Great question. It’s because the input directory is one folder up from the working directory.

[@sch0ll1]

@@robmulla thanks :)

[@ashbk1911]

Never used any of these for the final presentation in my 7 years of experience in data science. Power bi or Excel is the tool to go.

[@robmulla]

Wow. That is interesting. I've always though about those as more of data analytics tools. Reproducibility it key!

[@zeroheisenburg3480]

Nice video. How did you run the cell with the cursor staying active on the same position? Seems very useful. Edit: (Found the right keyword for search. It's Ctrl+Alt+enter) Damn, I've been using it for 6 months and didn't know about it. Could have saved a lot of time. This just shows no matter little things could be, it will always help someone out there somehow. Subscribed!

[@robmulla]

So glad you found out how to do this. It is really helpful to get all the keyboard shortcuts memorized for jupyter. Check out my video on jupyter where I talk about the ones I use the most!

[@nasirhameedkhan4630]

You are doing a great 👍job.

[@robmulla]

Thank you! 😃

[@marcgentner1322]

What is your editor called?

[@robmulla]

Jupyter lab. Check out my other video on jupyter.

[@philwebb59]

4:12 For some reason, with Pandas v1.5.3, "df_subset.columns = [c[1] for c in df_subset.columns]" doesn't split the multi-indexed columns like you show. I had to split the DataFrame by adding "['Count']" after the parenthesis when generating df_subset, otherwise you get "Id" values on the left and "Count" values on the right. Worse, the "Count" values are all colored zero.

[@robmulla]

Interesting. I thought I was using a new version of pandas in this video, but I'll keep that in mind in the future. Thanks for the comment!

[@blAnk7300]

I love the format, maybe use the same format for other things than visualization after it gets boring or repetitive

[@robmulla]

Great idea Anes. I'm trying to think of other options. Maybe modeling, but that's less visual and usually takes more time to train models.

[@co.n.g.studios5710]

Dfnitly wrong title, but an awesome vid on plotting!

[@robmulla]

Thanks. I appreciate that.

[@mehdismaeili3743]

Excellent.

[@robmulla]

Appreciate that Mehdi!

[@berkayberatsonmez5993]

Where can I find this wheel?🤔

[@robmulla]

Picker wheel dot com ?

[@alexeymatveev9031]

You are hero!

[@robmulla]

No. You are!

[@alexeymatveev9031]

@@robmulla 😆😆

[@darren7193]

Can you do one with a bank's balance sheet

[@arda3420]

5 minutes for each and on live stream could be exciting 🧐

[@robmulla]

That's a fun idea. Recording this look over an hour to edit down to 11 minutes! So I'm not sure I could make anything in 5 minutes, but maybe 15?

[@mahmodi5timetolearn]

A little fast teaching methods, but let me subscribe

[@mattvoss6673]

You’re clearly using pandas more than just importing the data.

[@robmulla]

You caught me 😅

[@clickbaitpolice9792]

Typo on the title. It should be “Royale” but you have “Royal”

[@robmulla]

Actually royal is the correct spelling. grammarist.com/spelling/battle-royal/#:~:text=A%20battle%20involving%20many%20fighters,battle%20royals%20or%20battles%20royal.

[@dakshbhatnagar]

How about you do this in 15 minutes or however many minutes you want (the time should be challenging enough) and the video has to be shot in 1 go. (no editing of video allowed during the coding time)

[@robmulla]

That's a great idea! I was tying something different for this video and wanted to edit it down so it wasn't too boring but also tried to not cut out any of the mess-ups. Doing it without editing the video would be hard! Maybe on a live stream? Thanks for watching.