A Gentle Introduction to Pandas Data Analysis (on Kaggle)
In this video you will learn the basics of how to use pandas in python for data science. Rob Mulla, kaggle grandmaster, will walk through the tutorial in a kaggle notebook. This pandas python tutorial for data science is great for anyone new to learning this technology. Pandas is a MUST for anyone getting into data science with python. Python is the #1 coding language for data science and has been growing over the years as an essential tool, with Pandas being the main data wrangling module. In this video we discuss the basics of how to use pandas including...
Timestamps:
00:00 Introduction
02:44 Series data structure
05:04 DataFrame data structure
10:01 Reading in data
12:17 Inspect The Data
15:25 Columns and Rows
19:00 Subsetting Data
25:45 Casting dtypes
29:00 Creating new column
31:19 Adding new Row
33:05 Plot Examples
36:45 Save out the output as a csv
Follow me on twitch for live coding streams: / medallionstallion_
Link to kaggle notebook used in the tutorial: www.kaggle.com/robikscube/the...
* KZhead: / @robmulla
* Twitch: / medallionstallion_
* Twitter: / medalliondata
* Kaggle: www.kaggle.com/robikscube
#DataScience #Pandas #Python #Coding
Timestamps: Introduction: 0:00 Pandas data structure: Series 2:44 Pandas data structure: Series 5:04 Reading in data: 10:01 Inspect The Data: 12:17 Columns and Rows: 15:25 Subsetting Data: 19:00 Casting dtypes: 25:45 Creating new column: 29:00 Adding new Row: 31:19 Plot Examples: 33:05 Save our output: 36:49
You are awesome @somuSan! Thanks so much for tagging these.
@@robmulla Nah no problem 😅, you are the awesome here, creating such a great end to end tutorial.
First pandas or python tutorial I was able to watched more than 10 min of in one sitting. Great job. Looking forward to following you here and on twitch
Hi there Rob! I wanna thank you for this fantastic tutorial you made on EDA with Python Pandas library. That really was the most impressive EDA tutorial I've ever watched on YT.
absolutely loved this video, one of the best walkthroughs out right now
Amazing introduction for pandas. This video alone just already covered most of my reporting needs.Thank you
I have been using pandas quite a while,learned a lot from your video. Your presentation was very good to follow through. Thanks for your hard work.
Really great! I love how you go through step by step and with the end to end examples. Thank you!
THank you so much for this video. The quality is amazing and you are such a good teacher.
Excellent intro. I had been toying with pandas but had not fully grasped the concepts of series and dataframes.
your intro i immediately was like you're my hero!
best I found on YT! underrated video, there should be more likes
First I want to thank you for sharing your skill and time creating these videos. I thought I was doing pretty well until we got to the Columns and rows section...haha I type the same "set_index" as you did but I kept getting an error. I found out I could run "reset_index", then I did "set_index" again and it worked.
Love it. Thank you Rob!
Awesome information, plainly stated
Thank you!! I never get tired of your videos!!!
I appreciate you watching them!!
This video deserves to be the most liked Gentle Introduction to Pandas Data Analysis on KZhead.
Really apprecaite that feedback. Share it with all your friends and maybe we can make it the most liked!
Wow, great video! Very much enjoyed it. I will definitely watch your other videos to learn many more techniques. Thank you.
You explain these in a very clear way,! Thank u sir 🙏
So cool! Pandas for Python makes things so much easier.
Excellent tutorial - Really clear with excellent explanations and concise steps. I've found it useful as a codealonng with a different set of data. Just to get to grasps with Pandas. Thank you!
Awesome! So glad you enjoyed and learned from it.
Thanks for the video, I've learnt a lot! I tried casting different dtypes on the imported Excel spreadsheet columns and found out an interesting issue, when I summed up the numbers in the same column with float64 and int64 dtypes, I got different results (no decimal points involved). Honestly I have no idea why it's working like that! I guess it has to be something with Excel formatting...
excellent video! amazing introduction! thank you a lot !
Thank you for the tutorial!
One of the best explanations thank you
Great video! Interesting way to make a Pandas Introduction. Thank you a lot to share with us your knowlegde!
Thank's for everything you share with us
My pleasure! I apprecaite the positive feedback!
Your channel is a goldmine ty!
Thanks!
Fantastic video! Thanks much for the lesson.
Excellent video, after 2 decades of experience in the software industry I cannot stress more how important it is to have a strong foundation in the basics inorder to attempt something more advanced. Wonderful job
Really appreciate your endorsement! I completely agree about knowing the basics first.
Interesting comment! I have been using python and pandas for years and thought to myself today I should go back and check the basics - so here I am! I find myself still googling every time I have to do simple stuff like change column order or make a copy of a data frame.
21:38 Wow!! I never knew you could use underscores in place of commas for the larger numbers!
just getting into DS and kaggleis great the notebook is so easy to use access to datasets is great. thanks for sharing!
Glad it was helpful!
Thank you for doing the video I learnt a lot about pandas please keep making videos like this.
I watched the whole video and I am so amazed at how easy is to do data analysis nowadays. I am the generation that used: Lotus to Excel, due to their limitation we needed to combine Statistical Packages(R, SAS), Graphing(Grapher), and 3D surface mapping (Surffer) plus Visual Basic. Now I have decided to learn Python plus others to go back to some old data for recalibration that was impossible deal with due to memory overloading. Panda is very powerful. Thank you very much for sharing your knowledge.
Glad the video showed you something new. Pandas is a great skill to master and can be really powerful for doing things that aren’t possible to do in excel with large datasets.
@@robmulla Yes, you are right. I am learning it and struggling with Jupyter Notebook and Kagle. I like both. I still have a long way to go!
Thank you very much, an amazing tutorial!!
That was really good ! Thank you !
Glad you found it helpful. Please share it with anyone else you think might also learn from it!
Amazing. Thank you!
yes, thanks a lot. l learnt so much. its been interesting and i am going to watch all you videos.
Amazing! Thanks!
Hey I think this is a great tutorial for actual pandas beginners. I have been using pandas for quite some time now, and still learned something. However, an advanced Video would be super great, because I see you doing all sorts of super advanced stuff in your coding videos that blow my mind and that I don't understand! Cheers
Thanks for watching. I’m glad you think it’s a good beginners! I need to think through what I could include for an intermediate/advanced tutorial. My pandas noob video has some good examples/tips if you haven’t already checked that out.
@@robmulla Hey thanks :) No I think I didnt watch it gonna check it out
@@robmulla I checked out the video and yes, I have learned a lot in this one! Also in the timeseries forecasting video. Now I feel like a noob haha
Although I watched this video a long time ago, I wanted to go over a basic explanation. Thanks for sharing your knowledge.
Glad it was helpful!
Got a new sub here. Thanks for your video.
Wow this was a great video!
Thank you Sir
watching your videos one by one bro, awesome work as usual
Thanks a ton, so glad you are learning from them. Make sure to comment and share so the videos get picked up by the youtube algorithm.
Great content!!
Amazing video, thank you Rob
Thanks for watching. What with a friend!
Thanks! So well explained.
Glad you enjoyed it!
For sure I enjoy this tutorial
Glad you did!
this is such a great tutorial, thank you so much!
Glad it was helpful! Please share with others who you think might also learn from it.
This video was so useful. Thank you so much!
You're so welcome! Thanks for watching.
Very well explained. easily understood . Didn"t have any issues following along. Good job, brother 😀😀
Glad it helped! I apprecaite the feedback.
You're very good at tutorializing.
🙏
thank you! greetings from Turkey :)
My pleasure. Glad to know there is someone watching from Turkey!
Great video
Amazing video!!!
Glad you liked it!!
please start a series of ML from scratch like linear regression, logistic regression, decision tree, random forest, k means clustering with python please sir please ..... i m really understanding your teaching style as you teach through hands on
Thank you so much.
Always welcome!
What is the shortcut you use @27:16 ??? To copy paste the code while replacing the variable inside?
Awesome 👍
Thanks Sandra! Glad you liked it.
thank you
Good content! 😄
Thanks so much Heitor!
Wonderful video - explanation is nice and clean, very intuitive narrative, thank you!
Thank you
You're welcome
Hello from dom rep next to aiti, how i change the back ground color of kaggle note book .
Hello, I saw one of your post in the youtube forum saying that your Timestamps didn’t worked out properly, I see now that it works. Did you do anything for it to work, any report sent to youtube ? Did the Timestamp worked after you hit the 1000 subscribers or the problem was there before ?
Hey! Yes. that was really frustrating. I didn't do anything special. It took a few months but then started working. I hope you get it figured out too.
Have a like good man. Gonna watch all of your videos from 0 ;)
Going on 1 full week of trying to change the font between Jupyter now Kaggle. Idk why I'm obsessed with it
Did you figure it out? Have you watched my tutorial about jupyter/python notebooks?
@@robmulla I have not. Adding it to the que now. Thanks!
Valeu!
For the last few weeks I've been struggling to learn how to do with Python what I do in 10 minutes with Excel. And a few months ago I spent half a day figuring out most of the steps to do the same analysis in Matlab. Dataframes are ridiculously difficult to deal with I'm learning it because several colleagues said that if I'm looking for a new systems engineering job, I should know Python. Numpy is easy. Pandas dataframes are obscure and clunky.
Interesting to hear you actually find numpy easier to work with than dataframes. I find the opposite because dataframes have an index and column names. It might just take some time and then it will click for you, or maybe you will find something else that works better for your use case. Hope the video was helpful and good luck on your journey.
Hi folks Does anyone know what keyboard he is using? Thanks in advance
At minute 12:40, I believe you meant first "5 rows" and *not* first "5 columns"
Ah! Good catch. Maybe I did it on purpose to make sure you were paying attention 😏
@@robmulla I truly appreciate the learning content you provide to the masses on your channel. Looking forward to more tutorials on Pandas, liked and subscribed!
great
thumb up. Thumb up
🙏
Getting errors while reading csv file
yass
😉
@@robmulla You just gained a new subscriber✅😊
Hello, What key do you use to open the menus?
Which menu is that? You might want to check out my video on jupyter notebooks when I talk about the keyboard shortcuts I often use.
What is I have an Excel or csv that has 999 rows but I want to delete 101 thru 999 rows? or Just read the first 100 rows only. Then save that as a new DataFrame as only those 100 rows.
.head(100) will give you just the first 100 rows.
@@robmulla Thanks for the info
Hi - why is the series of 'thing' in mydf dataframe an object dtype, but the 'count' is an int64 dtype? Why is 'thing' not a 'string' dtype?
I'm not sure, if "thing" is a string it will show as "object" if you count on it it will produce a number integer.. but I'm not sure exactly what you are asking.
@@robmulla hiya - thanks for the reply - at kzhead.info/sun/kqmbYNmsgISanq8/bejne.html you can see that the series containing stings has a dtype of object, whereas the series containing the integers has a dtype of int64. I was just wondering why its object and not string?
Very good 🙂👋👋👋👋👋👋😋
Thank you very much
Would be helpful to mention Tab kicks off the auto-complete for us n00bs
Good point. Also shift-tab for docstrings!
i cant convert the dtype of likeCount into int64 Can some1 hlp me
Hello Rob, Can you help me with this error, please? I can't figure out what i did wrong. Thanks a lot :) df['likeCount'] = pd.to_numeric(df['likeCount'].astype('str')) TypeError: 'Series' object is not callable
Strange. Don’t know why it isn’t working. Do you need to set the type to string? Try just dunking to numeric on the column without astype(‘str’)
@@robmulla Sorry again, I have a question, one more time. What is the best way to work with numbers with decimals, for example, I have a field that has decimal numbers, but I want to convert it to an int.
Can I use Google Colab to follow along also?
Yes! It's actually really easy. Just click the three dots in the top right corner of the kaggle notebook and then click "download code" you then can open it in google colab. You would also need to download and link the data. Why do you prefer colab over a kaggle notebook? They are very similar.
How many years have you been doing Data Science?
Thanks for asking. I've been working with data for over 10 years but doing data science specifically for about 6.
29:20 - it is giving me value error
do you have a discord channel??
Yes I do! Join! discord.gg/KnsDbstv
Dataframe 1 / Table 1 MaterialID |Unit Selling Price |Unit Cost A | 100 | 80 B | 200 | 140 C | 150 | 100 D | 250 | 230 E | 225 | 215 Dataframe 2 / Table 2 Month | Quantity Sold | Material ID Jan | 10 | A Feb | 5 | E Mar | 25 | C Jan | 5 | D Feb | 15 | B Mar | 2 | A Which month highest total sale amount achieved Which month highest profit amount observed Change Quantity Sold from 5 to 7 programatically and find out revised Which month highest total sale amount achieved Which month highest profit amount observed
My time stamps still not working -_-
Oh man. I’m sorry. That was really frustrating when it happened to me.
Don't use twitch. Will subscribe you here though
Thanks!
Hi Rob why are addressing Hey u tube ,seems to be quite odd or some sort , mind you U tube is not watching yours Videos we guys are watching our Likes and Views makes your day worth while , So Gentle request be direct to source please address us by " Hi guys " or something .. 🙏🙏🤞Any way Great Videos by You bro. Keep it Up. God Bless You...
Hello Rob, i followed your tutorial and i had an error when you created a new columns. Can you help me with this error. Thank you very much for the great job. /opt/conda/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead
Hey Carolina. This message usually comes up when you are trying to edit a subset of a previously defined dataframe. The best fix is to add `.copy()` after subsetting and renaming a dataframe. So before you might have: df_small = df.query('thing > 10') Change to: df_small = df.query('thing > 10').copy()
@@robmulla Hey Rob I am getting this error during converting the likecount into integer IntCastingNaNError Traceback (most recent call last) Cell In[126], line 2 1 df['viewCount'].astype('int') ----> 2 df['likeCount'] = df['likeCount'].astype('int').copy()