No video

End To End Text To SQL LLM App Along With Querying SQL Database Using Google Gemini Pro

  Рет қаралды 62,357

Krish Naik

Krish Naik

Күн бұрын

github: github.com/kri...
In this video we will be developing an end to end LLm application using Google Gemini Pro where we will create Text To SQL LLM App and later retrieving query from sql database
-----------------------------------------------------------------------------------------------
Support me by joining membership so that I can upload these kind of videos
/ @krishnaik06
---------------------------------------------------------------------------------------------------------------------------
►Data Science Projects:
• Now you Can Crack Any ...
►Learn In One Tutorials
Statistics in 6 hours: • Complete Statistics Fo...
Machine Learning In 6 Hours: • Complete Machine Learn...
Deep Learning 5 hours : • Deep Learning Indepth ...
►Learn In a Week Playlist
Statistics: • Live Day 1- Introducti...
Machine Learning : • Announcing 7 Days Live...
Deep Learning: • 5 Days Live Deep Learn...
NLP : • Announcing NLP Live co...
---------------------------------------------------------------------------------------------------
My Recording Gear
Laptop: amzn.to/4886inY
Office Desk : amzn.to/48nAWcO
Camera: amzn.to/3vcEIHS
Writing Pad: amzn.to/3vcEIHS
Monitor: amzn.to/3vcEIHS
Audio Accessories: amzn.to/48nbgxD
Audio Mic: amzn.to/48nbgxD

Пікірлер: 124
@krishnaik06
@krishnaik06 7 ай бұрын
Likes Target 1000 :)
@raghavendranarayana1009
@raghavendranarayana1009 7 ай бұрын
Likes target only 1000 that’s too small value mainly for your valuable content your Videos should get minimum 5k in a day
@yujibeballin4329
@yujibeballin4329 6 ай бұрын
Krish Bhai 729!
@036saurabhlasonkar8
@036saurabhlasonkar8 5 ай бұрын
Hello krish sir i have done all the code and execute that but the response is not working for me. I got correct Sql query but on streamlit it did not display anything can you help me
@BTSDaily7
@BTSDaily7 4 ай бұрын
Ho gaye bhaiya 👍👍👍
@karthikmurumalla3936
@karthikmurumalla3936 4 ай бұрын
Hi Krish! Lately I've been watching your videos on GenAi and I'm so glad that someone is putting up content on actually using these openly available models for real world use cases! However I would like to see more and more videos on domain specific fine tuning! Ive practically seen domain specifically fine tuned models with around 200M, 300M parameters generating results far better than generic models like Gemini, GPT3.5 turbo etc which are huge! I'm looking forward to a full length video from you on actually fine tuning smaller LLMs like T5 (base/small openly available) for domain specific tasks like text2sql and benchmark those scores with the results produced in this video!! Lighter models and more accurate results are always exciting to watch!
@BharatPatnawallah
@BharatPatnawallah 7 ай бұрын
You are great Sir learning stats and Eda from you. Very smooth explanation. 🎉🎉
@adikankwucharles9616
@adikankwucharles9616 7 ай бұрын
Nice one, sir. I am from Nigeria and got convinced in one of your videos to join Pwskills FSDSP batch. Please, I am still waiting for you. I need you in this industry. Thanks for all you do
@pdevender
@pdevender 7 ай бұрын
why dont you come to india
@adikankwucharles9616
@adikankwucharles9616 7 ай бұрын
@pdevender If I have an opportunity to work as a data analyst or a junior data scientist, I don't see anything that will deprive me of coming down to India.
@shivamjha9254
@shivamjha9254 7 ай бұрын
Thankyou sir🙏,from past 2 month I following your video.
@dario_real
@dario_real 8 күн бұрын
I'm a programmer and I love using all kinds of tools, especially those related to artificial intelligence. If you have a project, we can work together to make it a reality
@archowdhury007
@archowdhury007 4 ай бұрын
Wonderful video Krish. Beautifully explained. Keep posting.
@datasciencegyan5145
@datasciencegyan5145 6 ай бұрын
Krish mast banaya hai... I will try this in my ofc sample data
@BrijeshPatel-cd4yd
@BrijeshPatel-cd4yd Ай бұрын
Have you tried.. is it working with organization Tables. ??
@darshandaga7578
@darshandaga7578 7 ай бұрын
can you please put up a video on fine tuning this with a specific context? lets say i want to delevop this for a cilent. How would we go about then?
@myslates2854
@myslates2854 7 ай бұрын
Interestingly all the data science students got better marks than Devops. Possibly a Krish-AI Bias :)
@vamsitharunkumarsunku4583
@vamsitharunkumarsunku4583 6 ай бұрын
I have already mysql database with 20 schemas and each schema has multiple tables. In this case how to fine tune my mysql database with llm? Can you please explain. Thank you
@Munnu-hs6rk
@Munnu-hs6rk 2 ай бұрын
One idea is to Give the customize data in form of a er diagram and relational structure schema in form of pdf or something and fine tune the model
@TemporaryForstudy
@TemporaryForstudy Ай бұрын
hey what did you do?
@AayushSi7ha
@AayushSi7ha 7 ай бұрын
@Krish you are a legend.❤
@alexramos587
@alexramos587 7 ай бұрын
Thank you, Krish.
@debbh274
@debbh274 2 ай бұрын
Crisp and to the point...❤
@VictorAyalaSpooner
@VictorAyalaSpooner 5 ай бұрын
Thank you Krish, this video was great! Now I have a question, I am an app developer and I use database services such as Firefox or Parse Sever, do you have any idea how to approach the same with those databases?
@musicphimhan7530
@musicphimhan7530 Ай бұрын
Thank you for your lecture.
@miteshgarg9420
@miteshgarg9420 6 ай бұрын
Hi Krish, thanks for the amazing video. Can you please help to create a similar solution using open source llm models like codellama. How can we fine tune the open source llm model to work like text to sql solution
@user-ds3qc8ge4k
@user-ds3qc8ge4k 5 ай бұрын
Hi Krish, thanks for the amazing video. Can you please help to create a similar solution using llama2 models . How can we fine tune the open source llm model to work like text to sql solution
@Iqbalshah-fd8yk
@Iqbalshah-fd8yk 7 ай бұрын
You are wonderful sir
@yuvrajdawande6382
@yuvrajdawande6382 7 ай бұрын
Just wow 🌟
@karthikbalasubramanya
@karthikbalasubramanya 7 ай бұрын
Ultimate...thanks krish
@shashankpandey1966
@shashankpandey1966 7 ай бұрын
I had made one using Hugging face .
@pdevender
@pdevender 7 ай бұрын
could you please share your hugging face space link
@bihanbanerjee3541
@bihanbanerjee3541 6 ай бұрын
could you please share that link
@haku660
@haku660 5 ай бұрын
Hi krish Sir , i wanted to show the Query on the front end so basically Search(Question) -> Sql Query should be the answer
@somethingnew7979
@somethingnew7979 7 ай бұрын
This video is really helpful for me ,sir I have one doubt can we use ready made data
@wassimdiai6
@wassimdiai6 7 ай бұрын
thank you for the useful informations you provide in all of your videos, i have a question regarding this video , can we add a performance metric to make sur the query is correct
@allaboutworld5177
@allaboutworld5177 7 ай бұрын
Sir is there anything like we can upload pdf then the text in pdf should translate in another lang and the translated text should replace the text in pdf with same format of pdf and generate translated pdf if possible then sir please let me know.
@meghanaarvapally5484
@meghanaarvapally5484 5 ай бұрын
Nice idea bro...😊
@user-po5rg8zu6e
@user-po5rg8zu6e 5 ай бұрын
Made my life simple
@rvraghvender
@rvraghvender 7 ай бұрын
I think the information posted in the "JOIN" button is quite old. Please re-explain what are the benefits we get in different subscription option.
@krishnaik06
@krishnaik06 7 ай бұрын
all newly fresh videos on ML and DL are getting uplaoded. Check the below playlist kzbin.info/www/bejne/n6GXiKt7j9x4hsU
@Shivang0369
@Shivang0369 4 ай бұрын
Great Video Dear Krish Naik Sir :") Can I use this approach for larger projects ??? Like i have HRMS project which has so many tables and data , can you please answer to my question wether this approach is suitable or I have to use different approach ????
@divyanshdhadwal9530
@divyanshdhadwal9530 Ай бұрын
Sir how can we add another table in db and join them and on the basis of that joined table how will our model respond. Like in the given table there are name, class, subject and marks. In second table there is name and their roll number. So how can we join these two tables and make our model train over it please help
@selub1058
@selub1058 Ай бұрын
In the same way. It doesn't matter how many tables you have in the database. Just adjust the prompt a little.
@wheresthecode
@wheresthecode 5 ай бұрын
would be nice to see this working with llama and not gemini only. Please show us how to use llama
@satbhattacharya6481
@satbhattacharya6481 2 ай бұрын
Excellent video!
@guyknowsanime
@guyknowsanime 5 ай бұрын
sir ap aik application bna de jesse apne q/a based bnai thi streamlit or gemini api key k sath. vhi cheez fastapi k through bnani ha
@gerokatseros
@gerokatseros Ай бұрын
actually last query was wrong, it would return all marks less except from largest. The question was to find student with second mark.
@srikrishna6595
@srikrishna6595 7 ай бұрын
Can you please train more data to the SQL server
@mohamedelzonkoli6680
@mohamedelzonkoli6680 7 ай бұрын
What is the best way to have an offline LLM interacting with sqlite db?
@Ai-ni3gs
@Ai-ni3gs 7 ай бұрын
Sir new video to store data into vector database free tier….
@datascientist_
@datascientist_ 7 ай бұрын
Best video.
@lenkamndev9988
@lenkamndev9988 7 ай бұрын
Sir could you post a video on your time management.
@MuhammedSavadkv
@MuhammedSavadkv 7 ай бұрын
Great😊
@Abhishek_Aman06
@Abhishek_Aman06 7 ай бұрын
How can we get only query out of this chain ?
@palashjain4203
@palashjain4203 4 ай бұрын
Krishna, great content! Can we use this for generating pictorial output representation.
@Debanjali-uk7wf
@Debanjali-uk7wf 7 ай бұрын
sir my API Key is not generating
@shivanisharma9554
@shivanisharma9554 6 ай бұрын
I tried it with one database table with timeseries data in postgresql, It is giving entire records no matter what question I ask?
@programmerpro-hs8ro
@programmerpro-hs8ro 5 ай бұрын
Did you resolved it
@08inside
@08inside 7 ай бұрын
How can we scale it for real world data with 100s of tables/columns? Any refernces?
@simply54
@simply54 7 ай бұрын
even i had the same query ? can we have multiple tables ? as in this video he has used a single table
@sahilsaini4574
@sahilsaini4574 6 ай бұрын
try rag
@brightlyricsmusic
@brightlyricsmusic 7 ай бұрын
Could not create `Blob`, expected `Blob`, `dict` or an `Image` type(`PIL.Image.Image` or `IPython.display.Image`). Got a: Value: ['You are an expert in converting English questions to SQL query!The SQL database has the name STUDENTS and has the following columns NAME, CLASS, SECTION and MARKS For example, Example 1 How many entries of records are present?, the SQL command will be something like this SELECT COUNT(*) FROM STUDENTS; Example 2 - Tell me all the students studying in Data Science class?,the SQL command will be something like this SELECT * FROM STUDENTS where CLASS="data science"; also the sql code should not have \'\'\' or any quotes symbol in beginning or end and sql word in the output. '] getting this error someone help
@programmerpro-hs8ro
@programmerpro-hs8ro 5 ай бұрын
Did you resolved this one buddy?
@brightlyricsmusic
@brightlyricsmusic 5 ай бұрын
@@programmerpro-hs8ro join all the prompts in single string using " ".join
@alaparthinagabhaskar2318
@alaparthinagabhaskar2318 3 ай бұрын
pass a string in the method instead of list like "model.generate_content(prompt[0] + question)" this will resolve the above error
@KumR
@KumR 7 ай бұрын
Can this connect to Oracle Database?
@KokkeOP
@KokkeOP 7 ай бұрын
im interested in this too.
@abhijitpatra1387
@abhijitpatra1387 3 ай бұрын
in out put i am getting paranthesis and single quotes so how to remove it
@krishnaprasadsheshadri6206
@krishnaprasadsheshadri6206 6 ай бұрын
damn this is crazy, what a great video. How would one go about scaling this application for a larger dataset and maybe for different kinds of data like geospatial etc?
@n.aniruddhan4464
@n.aniruddhan4464 7 ай бұрын
Sir NLP with deep learning one shot please
@DineshkumarRamasamy
@DineshkumarRamasamy 4 ай бұрын
Have tested with one table and it works all good. But struggling with multiple tables Can you provide a sample with multiple table in one database ?
@Imvishnuch
@Imvishnuch Ай бұрын
Hi krish, we have this already but the query is taking 1 - 2 mins, i am an AI product manager, can you share techniques to reduce that time to 5-10 seconds ?
@swatikeshari858
@swatikeshari858 4 ай бұрын
If I want to read and ask questions from multiple csv files at a time, then how can I implement?
@anirudhgrack7560
@anirudhgrack7560 7 ай бұрын
Google AI Studio not available here in canada
@anirudhgrack7560
@anirudhgrack7560 7 ай бұрын
Any other way out ??
@swatik2509
@swatik2509 7 ай бұрын
​@@anirudhgrack7560leave canada
@bellosaburoladipupo7430
@bellosaburoladipupo7430 6 ай бұрын
Hello Krish, how do I fix a 'DefaultCredentialsError'? I can't seem to get the app answering questions despite inputing my Api key. Thanks
@_csds_PranjalSharma
@_csds_PranjalSharma 3 ай бұрын
plz make video on how to deploy it
@DataPulse
@DataPulse 7 ай бұрын
Krish the king bhai
@arinspace6588
@arinspace6588 4 ай бұрын
Hey Krish, amazing work. I have been trying to run this on an AWS Linux instance and it throws me (argument of type 'Part' is not iterable) error. It works fine on Windows local PC. Can you help me fix this issue please.
@arielgamao5480
@arielgamao5480 4 ай бұрын
Hi, do you have videos for chatbot? thanks
@imranurrehman264
@imranurrehman264 7 ай бұрын
great,,,
@satyajamalla1429
@satyajamalla1429 6 ай бұрын
Can I have a llm for Tex to BDD test cases for java applications
@1984tushu
@1984tushu 2 ай бұрын
Can it also run on any warehouse SAP BW, Synapse etc
@chihebnouri5541
@chihebnouri5541 5 ай бұрын
can you amke a video how to use that in angular?
@datasciencegyan5145
@datasciencegyan5145 6 ай бұрын
Is it possible to retrieve table name from where ever it in any schema with respect to column name
@abdulrahmanalrabah8324
@abdulrahmanalrabah8324 3 ай бұрын
Would it be possible to have it sql to txt instead ?
@JayantKumar-lq9lg
@JayantKumar-lq9lg 5 ай бұрын
Can you pls make video like same for NoSQL db by using Mongo
@dulshansamarakoon5068
@dulshansamarakoon5068 7 ай бұрын
Wow..
@user-xj6bx1hn5s
@user-xj6bx1hn5s 7 ай бұрын
Thank you , was very informative . How should I go about implementing if the data model is complex with 10 tables and has star schema ?
@mithunshet5922
@mithunshet5922 7 ай бұрын
This is not reliable.. llm should be used to correct the SQL code if at all an error exists.. But this questionnaire is unreliable for complex data models..
@KokkeOP
@KokkeOP 7 ай бұрын
@@mithunshet5922 Honestly, what he should do is simply try his .db and see if it retrieve the right results with his queries. its not hard as long as the code is ran from his on-prem/local environment, the results should output a plausible answer. The hardest part should be to update the Prompt to the right kind of questions the model is going to be enquiry.
@protimaranipaul7107
@protimaranipaul7107 3 ай бұрын
Have you found a way? Multiple agent?
@tharunps8048
@tharunps8048 7 ай бұрын
I have a doubt. Will this send my Database contents/schema along with API request to LLMs for queries I ask ?
@luckytalkz9442
@luckytalkz9442 5 ай бұрын
Sir when enter Streamlit run app.py mine display comes empty. Why? I am very confuse.
@luckytalkz9442
@luckytalkz9442 5 ай бұрын
Hello sir please tell me? Why this happens
@BharatPatnawallah
@BharatPatnawallah 7 ай бұрын
Krish Sir I am from sales background but wanted to switch started learning but not sure how to proceed and land a job in the end. Please make a session for which we can increase our chance to get interview calls. Is there any mock preparation or Hr connect available in ineuron
@raghavendranarayana1009
@raghavendranarayana1009 7 ай бұрын
I am also from Marketing background I moved to Data science in 2022 if you understand the basics and statistics ML,DL,CV and NLP is nothing
@BharatPatnawallah
@BharatPatnawallah 7 ай бұрын
@@raghavendranarayana1009 which domain have you chosen for your career
@pdevender
@pdevender 7 ай бұрын
@@raghavendranarayana1009 right now are you in ML domain. How did learn this things. Did interviewer concerned about your background
@raghavendranarayana1009
@raghavendranarayana1009 7 ай бұрын
@@pdevender interviewer asked on why did you transit from marketing domain to Data science domain
@pdevender
@pdevender 7 ай бұрын
@@raghavendranarayana1009 ok what was your answer then ?
@yvens_data
@yvens_data 6 ай бұрын
Hi krish , I tried to push it to github and deploy the app on streamlit community, the data base stopp working, how can i fix that
@programmerpro-hs8ro
@programmerpro-hs8ro 5 ай бұрын
Did you complete this project?
@jawadasif8130
@jawadasif8130 6 ай бұрын
how we can use Gemini in EU? Anyone got the Idea?
@qzwwzt
@qzwwzt 3 ай бұрын
Do I have to change thw peopt language? I mean, may I wrote it in Portuguese, for example?
@TheIITianExplorer
@TheIITianExplorer 3 ай бұрын
Sir it depends, I think Gemini can understand Portuguese... So u can write in Portuguese
@shashankpandey1966
@shashankpandey1966 7 ай бұрын
First❤
@bangarrajumuppidu8354
@bangarrajumuppidu8354 7 ай бұрын
Second ❤
@programmerpro-hs8ro
@programmerpro-hs8ro 5 ай бұрын
Anyone did this one?
@rishichowdhury4296
@rishichowdhury4296 7 ай бұрын
@krishnaik can we use my SQL instead of SQL lite ?
@krishnaik06
@krishnaik06 7 ай бұрын
Yes you can
@RobbieMraz
@RobbieMraz 4 ай бұрын
Hi do you know, if Let's say we connect Azure OpenAI to a SQL DB (Oracle) and want to retrieve data using AI Azure Platform, will SQL DB have an impact performance if so do you know how much can be impacted? Let say that we use SqL dB with many integrations
@rohithak9419
@rohithak9419 7 ай бұрын
Like number 674
@nileshbhujbal7489
@nileshbhujbal7489 Ай бұрын
I will tell your AI bot to delete all data from xyz table...👿
@MUsmandev
@MUsmandev 7 ай бұрын
whatsApp bot with gemini pro is this possible ............?
@abhishekfnu7455
@abhishekfnu7455 5 ай бұрын
Is there a way to use Data Dictionary to train LLM model to generate SQL queries later on?
Chat with SQL and Tabular Databases using LLM Agents (DON'T USE RAG!)
58:54
а ты любишь париться?
00:41
KATYA KLON LIFE
Рет қаралды 3 МЛН
Why Is He Unhappy…?
00:26
Alan Chikin Chow
Рет қаралды 110 МЛН
How I Did The SELF BENDING Spoon 😱🥄 #shorts
00:19
Wian
Рет қаралды 36 МЛН
Joker can't swim!#joker #shorts
00:46
Untitled Joker
Рет қаралды 39 МЛН
A Natural Language AI (LLM) SQL Database - Could this work?
8:52
Discover LlamaIndex: Joint Text to SQL and Semantic Search
20:08
LLM for data analytics: text-to-sql 3 architecture patterns
6:59
Denys on Data
Рет қаралды 1,3 М.
LangChain Crash Course For Beginners | LangChain Tutorial
46:07
codebasics
Рет қаралды 281 М.
I Studied Data Job Trends for 24 Hours to Save Your Career! (ft Datalore)
13:07
Thu Vu data analytics
Рет қаралды 191 М.
Chat with MySQL Database with Python | LangChain Tutorial
37:11
Alejandro AO - Software & Ai
Рет қаралды 43 М.
How This New Battery is Changing the Game
12:07
Undecided with Matt Ferrell
Рет қаралды 207 М.
а ты любишь париться?
00:41
KATYA KLON LIFE
Рет қаралды 3 МЛН