Analysing Data with ChatGPT (Data Analysis and ML )

In this tutorial we will see how to analyse a given dataset using ChatGPT.

⚡ Analyzing Data with ChatGPT⚡

🎓=== Check out these Awesome Data Science Courses!===🎓

⏲️===TimeStamps===⏲️
0:01 Introduction
00:45 Analysing Datasets
01:20 Convert Data into a Table Form
03:30 Class Distribution and Pivot Tables with ChatGPT
06:02 Value Counts
06:45 Group By with ChatGPT
08:01 Encoding Dataset with ChatGPT
09:25 Model Building
11:20 Evaluate the Model
12:10 Model Interpretation
15:08 Recap

If you liked the video don't forget to leave a like 👍 or subscribe ❤️.
⚡ If you need any help just message me in the comments, you never know it might help someone else too. ⚡

J-Secur1ty JCharisTech

Support the Channel: Become a Patreon

◾◾◾Get The Data Science Prime App◾◾◾

◾◾◾ Need Your Dataset Cleaned check out this gig ◾◾◾

Follow

[BingLi]

Be careful. Double check the accuracy of all messages from ChatGPT.

[Shields]

I love how excited you get even when you know what the outcome of your own tutorial is going to be. Keep it going!!

[Sad_Bumper_Sticker]

Your excitement is like the high intellectual engagement of a scientist or a researcher. It makes watching / listening engaging.

[arr]

I am a data analyst/data scientist for survey analytics. I have found that althought AI can perform many of the function I require for producing the models and tables that I program, it does not do it much faster than my current methods. For simple surveys it is possible to feed the data into these programs and retrieve simple output, however, when you get into more complex processes such as data layering of large data sets and merging new data etc, it often produces results that are difficult to troubleshoot. Meaning, you must be an experienced programmer to use AI to produce the same results that I am already creating using other software. I find that it appears to save a bit of time initially, but that checking your work becomes more risky. MY current programs are easier to see the data you are manipulating which would be much more difficult to keep track of in this format. I do not think that even if AI can handle large data sets, that it is going to produce the results much faster. There are far too many subjective based decisions that need to be made on data sets in order for this porogram to replace these traditional methods. Most of my job is actually data management and when you get to the point where the data is ready to be analyzed it is already a very quick process to gather these results. It is possible that I may deploy AI in addition to my regular routines in some circumsatnces but I do not yet see the practical applications of switching over to any large extent. I beleive this will be true for most of the jobs that AI threatens to take over. You still require someone who is experienced to manipulate and interpret the data. I am excited at the potential for AI to aid in translating these findings into a report in order to save time with the actual analysis. I belevie it will allow for more thorough projects rather than rendering these jobs obsolete. So far, I have found Chat GBT very useful in guiding my ideas around how I create models since it can reason whether or not my logic seems reasonable. It has also helped me to learn new techniques and expand my offerings but again, I do not see it replacing or lessening the value of my work. If it is able to get to the point at which is replaces my job entirely, I beleive that the entire industry itself will be replaced.

[nafisasadia]

This video is a life changer for an aspiring data analyst/ scientist with minimal coding skills; you could literally create a whole course on using chatgpt for data science and analysis with demo use case examples like this one for all sorts of data science process workflows for all types of cases. It could make your channel famous as this is very new and for me, launch my data science career with much ease.😅 But anyways, amazing video brother 🎉

[niundisponible]

Because If you can notice, at the start of creating the table, there are some | characters denoting column separation, and "-" character denoting separation. This is when you, for example write code in Jupyter format, and according to the characters and the format it will display as a "markdown", which can interpret different syntax to show tables, equations, code snippets, lines of codes, hyperlinks and some more stuff. This is how you make a table in markdown syntax:

| Name | Age | Gender |

| John | 25 | Male |
| Jane | 30 | Female |

And Chat GPT can interpret this syntax, as it is available on the internet. The display window of ChatGPT's output responds to markdown syntax and displays the text as either a code snippet or a table. It is still impressive though!

[RahilKasimi]

You can ask ChatGPT to 'complete or finish above code' if it gives only half the result.

[HamsterMazeAdventure]

ChatGPT is so impressively powerful!
.. can't wait to see GPT-4 in action! ✨

[eliotharreau]

Formidable ! This is the next level of the humanity...

[streetchronicles]

I was kinda cursing at this dude at how basic all this was up to 7:40ish lol... very impressive, awesome video. I had guessed what you were going to say, but I thought "give it ranges, 0-300, etc, but what you said and what it did was extremely impressive.

[cryptosfool]

Great presentation. I loved how you present and your excitement. Like you I find amazing how it is able to infer the data meaning and columns. I gave it a list of zipcodes and district names and it inferred correctly that the dataset was a list of zipcodes in Puerto Vallarta Mexico. Simply amazed.

[AndrewCatanach]

Thanks, an excellent, concise presentation, love the passion! I agree with other comments to be skeptical of output of data analysis by ChatGPT but you've shown how it can be a great teacher for now.

[digitalgeek]

This is mind blowing. Great presentation. It also bores down to the user knowing the right questions to ask

[madikargin]

The last bit : "eto ochen' interesno" just made my day 😁

[gxhkgtk]

Clever!! Your video saves me a lot of trouble.

[bane]

cool vid. I wonder what the limit is for the amount of data gpt will process

[RichardHamnett]

When the model stops, just type "continue"

[cashbuyer]

I always start off using chat gpt, which quickly turns into me doing it myself to check chat gpt accuracy which makes it kinda useless for things you need to be accurate. I still love prompting it though!

[BrendanClements]

It's not stopping because it can't continue. It stops because you're using a test window instead of the API.

[DjehutimasAsarRa]

Game changer. Just type continue when it stops in the middle of generating code.