PandasAI - Data Analysis Made Easy (Powered by OpenAI)
ฝัง
- เผยแพร่เมื่อ 28 มิ.ย. 2024
- To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/ThePyCoach/. The first 200 of you will get 20% off Brilliant’s annual premium subscription.
In this video, we'll learn how to use PandasAI, the new data science library that integrates generative AI capabilities into Pandas.
Github: github.com/gventuri/pandas-ai
Code & Data: github.com/ifrankandrade/pand...
🔥 My FREE ChatGPT Cheat Sheet: frankandrade.ck.page/08c94cf1c1
My Free Cheatsheets
==================
🤖 Python Automation: frankandrade.ck.page/44559e1ae7
🐍 Python for Data Science Cheat Sheet: frankandrade.ck.page/26b76e9130
📐 R for Data Science: frankandrade.ck.page/7b621bc82c
🛢️ SQL for Data Science:frankandrade.ck.page/f2fbbdc6db
🌐 Web Scraping: frankandrade.ck.page/ca38420833
📘 My eBook “Code Less Earn More”: frankandrade.ck.page/3b520320ac
Social Media
===========
🎮 Discord: / discord
🔗 LinkedIn: / thepycoach
🐥 Twitter: / thepycoach
✏️ Medium: / frank-andrade
😺 GitHub: github.com/ifrankandrade
My Courses
==========
🔥 Join My Automation Course in Python: www.udemy.com/course/automate...
🔥 Join My Python for Data Science Bootcamp: www.udemy.com/course/python-f...
🔥 8-hour Web Scraping Course in Python: www.udemy.com/course/web-scra...
💰 Make money by writing about AI, programming, data science or tech: thepycoach.teachable.com/p/me...
Support My Work
==============
💵 PayPal: www.paypal.com/donate/?hosted...
Disclaimer: This video was sponsored by Brilliant.
![[LIVE] : ONE ลุมพินี 68 | คู่เอก "พระจันทร์ฉาย vs โจนาธาน"](http://i.ytimg.com/vi/F3IM3w-XchI/mqdefault.jpg)
To try everything Brilliant has to offer-free-for a full 30 days, visit brilliant.org/ThePyCoach/. The first 200 of you will get 20% off Brilliant’s annual premium subscription
Thanks for this video. I´m working a lot with pandas, so this is definitively a new way to process data. Amazing to see how AI can make our lives (in programming) easier.
In your prompt for the bar plot, try having it work in steps.
"Calculate the total spent by each gender. Plot those numbers in a bar plot."
Phrased that way, see if you get the correct plot.
I tried your tutorial today, and found they fixed this problem.
I have this same problem with chatgpt 4 in just LLM form. Sometimes it makes up numbers. I think if you are more specific like “create a bar plot based on the previous calculation” it will be more accurate.
I got this error: ImportError: cannot import name 'PandasAI' from 'pandasai'. How to solve it pls?
Promising idea👏
Excellent!
Thank you for this video - I wonder whether it is possible to use pandasAI to extract data from multiple documents and store them in an editable series of form fields such that a user will be able to speed up the process of data extraction...
This is very cool!! Have learned a new or better method to chat with data?
probably will be something similar in the future with other libraries like scipy numpy etc
thanks very helpful.❤
can use local hosted llm like TheBloke/Pandalyst-7B-V1.1-GGUF as llm for pandas ai instead of openai ??
please make video ??🙏
What is the difference between this and ChatGPT's noteable plugin?
i have an stock data, while sum the stock qty, i want to apply some number to char convertion funtion how to do... in pandas ai
EX: stock qty=1111
i want a result 1 ctn/1pkt/1 box/1 pcs like this
Pandas AI looks great. I wonder if you were using the GPT4 API it would have made that mistake. I have found GPT4 a massive improvement over chatGPT and wonder if you'd see that improvement here too. And also how it compares to Interpretor in chatGPT. And I assume you can use it with the other LLMs on Hugging Face? Exciting times, especially for those of us who don't know much about coding.
Good question. I believe the account I'm using doesn't have access to GPT-4. Yeah, you're right you can use it with Hugging Face too
How much it $ cost everytime you run a code whith the api?
I just start using Langchain. Is pandaai better?
5:50 That's weird, it looks like its only totaling the first five rows of the dataframe. I wonder if you'd get the correct results if you specified to use the entire dataframe in your prompt
how to slove this error
UnicodeEncodeError: 'latin-1' codec can't encode character '\u200b' in position 7: ordinal not in range(256)
When I use PandasAI and give it the dataframe, is that data being sent to openAI servers?
hey Although I am using Paid API key, the API is taking long time to respond. is there any chance , we can improve the Performance
Is PanadasAI stable? Not getting as expected
Is this using the free or paid account from OpenAI?
I keep getting this error
"data type 'Int64 not understood'
I have tried to search solution in pandas ai git but still can understand solution. Please help
Is there anything similar to pandasAi
I think the problem is related to conversational memory maybe?
Not quite sure. Hopefully they'll fix this soon
I get this error: This model's maximum context length is 4097 tokens
How can i solve it?
you could try using word embeddings or langchain
1st time I use the API key but in VSCODE after running my code showed this "You exceeded your current quota, please check your plan and billing details." Does anyone help me.....🙏🙏🙏🙏
It might be the first time you use your API key but probably your free trial have already expired. I think it started the day you created your OpenAI account. You have to either create a new account or add a payment method
@@ThePyCoach Boss I there any other methods to use API or get API ???
Tell me my friend also using ChatGPT but their API working fine but mine not
Is it okay to display your token publicly like that?
It's not. That's why I deleted the token after making the video 😉
Should send the code back too
how to buy Tokens ? i am getting the following message
"Unfortunately, I was not able to answer your question, because of the following error:
You exceeded your current quota, please check your plan and billing details.
"
I figured it out ...
@@parikannappan1580 how?