hello, im new in this channel... and programing XD, and I need help, im triying to make a website with flask, html and css and I have it all ready, Im using sqlalchemy and and the website works perfectly... in local, i dont know how to make it work in a server, its an ubuntu server and you maybe already explained this but as I told you before im new in all this so if coul give me some advise i would be gratefull
when i try 34:49 i get some information that does not exist in the csv file, on top of that, i get some columns that don't even exist int the csv file like this : [Shipping Method: Standard Ground Shipping Address: [Your Address]] i'm using llama3-70b-instruct. is this a problem with the model ? because i tried a lor of nvidia models and they either say i'm a conversational model i can't do this, or i recieve an error saying the model doesn't support what i'm trying to do
Man this is dense and excellent, no time wasted! Really good example. I have to look at the security side of this solution. Thanks a lot for producing great content.
This was worth every minute to work through. Thank you for making it so easy to understand, and setting a solid foundation to build from. You are an excellent communicator!
Thank for all the hard work you put into these tutorials. I’ve watched thousands of hours of worth of tutorials on TH-cam and you are among the best. Keep up the hard work.
idk if you have answered this already but i guess you are changing the agent instruction instead of the description @29:23 so since you changed the instructions again later description is default. Thank you for your the video you really did a great job😊
For those having problems with the AI not answering correctly or just saying I don't have information even though it is in the db whatever you are asking then do this. So for the Manager Agent controls, there is Agent Description, make sure to turn on the toggle so it's easier to edit and then write there: A helpful assistant with access to the following tools: 1. FAQ 2. OrderLookupAgent Replace with whatever names of the Agents you have, and everything should work fine. Do this for other Agents as well, if their description is empty, so just write something.
Great tutorial. Thank you for the detailed update from the previous langflow episode, since they updated so soon. I followed everything and got a slightly different test response from the frontend: "I'm currently unable to provide specific details about the cancellation policy. However, generally, most companies allow cancellations within a certain timeframe after placing an order, typically before the order has been processed or shipped. If you have a specific order in mind or need assistance with a cancellation, please let me know!" I have all of the data setup in my database as followed in your video. Do you know why this happened and how it can be avoided?
been trying for some time and cant get it to show any product or order data - always reports it does not have access to it. Followed step by step a few times and same results. I wonder if something has changed in langflow that requires an additional step? Or am I missing something - others report the same thing in the comments
@@Mellybontest @theonlytapy @wakeboardr886 @Moths-m2r - ive not solved it yet no, kind of had to leave it for now as didnt have time @TechWithTim - Would be good if you have any ideas Tim? Great video as always BTW
Hello, please enable the option for Portuguese (Brazil) in your videos. The community here will be happy and eager to follow you. Thank you, and congratulations on the great video!
Hi Tim, great job on this! I need your advice on debugging the flow where the system accesses the "products" and "orders" tables. My flow is identical to yours but isn’t retrieving any data from the tables. I noticed that disconnecting the FAQ module and ManagerAgent resolves the issue, and the "products" and "orders" responses work as expected. It seems the ManagerAgent has trouble handling results from both tools. The same issue occurs when I import your JSON file into LangFlow. Appreciate your help!
I was facing a similar issue. In my case, it was because the field "Astra DB Application Token" in Astra DB tools for look ups was incomplete. I don`t know why it didn´t select ASTRA_DB_APPLICATION_TOKEN by default, but anyway, once I completed this it started to work.
I'm getting this error could you please help ValueError Details: Error performing search in AstraDBVectorStore: Ordering on non-clustering column query_vector_value requires the column to be indexed
I encountered the same issue. In my case, my PDF initially consisted of plain text without proper titles, or visually structured paragraphs. To address this, I later added titles for each paragraph, organizing the content into questions, as shown in the video, and my issue was solved. I wish you are able to resolve that problem too.
I was able to fix my error, but unsure if its the same as yours. I went to customer overview, then on the right hand side I generated a new token, and updated my astradb token key. It seems to work now. It is also possible it was just on the datastax side.
when i try 34:49 i get some information that does not exist in the csv file, on top of that, i get some columns that don't even exist int the csv file like this : [Shipping Method: Standard Ground Shipping Address: [Your Address]] i'm using llama3-70b-instruct. is this a problem with the model ? because i tried a lor of nvidia models and they either say i'm a conversational model i can't do this, or i recieve an error saying the model doesn't support what i'm trying to do
Great video, thanks a lot !!! Is it possible to use a csv instead of a pdf? I would like to download a csv containing a table from a database to obtain answers from the questions asked (chat-to-csv)?
Sir, I have tried using the Ollama model and need to connect it with the agent to access the tools. However, I am unable to do so. Is there any possible way to connect Ollama with the agent? Thank you.
Hi! I have a doubt related to the flow of the AI app. Is it really necessary to drag and drop the Vector Database again when loading the PDF file? I mean, I passed them to Ingest Data of the existing Vector database, and it seems to work. I'm not sure if I'm missunderstanding something. Thanks a lot :)
I'm trying to build the same thing, but instead of OpenAI, to use Mistral AI - but it does not work - is it enough for Agent to select Other language moel (not OpenAI) and link Agents Language Model pink dot to Mistral AI Language model pink dot - or does it need some other connection as well from LLM to Agent?
I used Google Gemini. I added the Google Generative AI under Models. I made my Agent Language Model is custom. Then connected the model to agent by Language Model
i dont know when somone is following all the exact steps like you do and still getting errors how is it possible like i feel all that you did was false, like someone cant get the result even though doing same steps i am frustrated now on some stupid error damn
How is this agentic Ai different from Generrative AI, In gen AI we have aa asimilar project we can make a chatbot which can give ans to multiple pdfs at same time. Isn't this project similar to the one taught in this video? Plese explain anyone
I'm getting this error after trying to run the multi-agent in the playground: Error Building Component Network error. Please check the connection to the server Can anyone help?
This is literally the best way to pay to win. You buy an extremely good computer, and your enemies will be stuck with 10fps on a normal machine (I guess the fps drop is also on the machine of the enemies
Get Started with Langflow: dtsx.io/agents
Thank you for your wonderful videos Mr.Tim.
I love your videos.i am very much interested to gain more knowledge from you.
😊
hello, im new in this channel... and programing XD, and I need help, im triying to make a website with flask, html and css and I have it all ready, Im using sqlalchemy and and the website works perfectly... in local, i dont know how to make it work in a server, its an ubuntu server and you maybe already explained this but as I told you before im new in all this so if coul give me some advise i would be gratefull
thx for link tim❤❤
when i try 34:49
i get some information that does not exist in the csv file, on top of that, i get some columns that don't even exist int the csv file like this : [Shipping Method: Standard Ground
Shipping Address: [Your Address]]
i'm using llama3-70b-instruct.
is this a problem with the model ? because i tried a lor of nvidia models and they either say i'm a conversational model i can't do this, or i recieve an error saying the model doesn't support what i'm trying to do
Man this is dense and excellent, no time wasted! Really good example. I have to look at the security side of this solution. Thanks a lot for producing great content.
Would you post what you find regarding the security?
yah. post your finding or at least make a blog to share your knowledge.
Dude, this is freakin' great! I can't believe this video hasn't gotten over 100k views yet.
This was worth every minute to work through. Thank you for making it so easy to understand, and setting a solid foundation to build from. You are an excellent communicator!
Just started and I love it already. thank you for this great tutorial
you getting better and smarter...blessed brother
Thank for all the hard work you put into these tutorials. I’ve watched thousands of hours of worth of tutorials on TH-cam and you are among the best. Keep up the hard work.
Yeah dude same here, watched so many tutorials but the clarity Tim offers in his videos for beginners is better than 999 other tutorials
awesome tutorial!
want more end to end projects with langflow and streamlit/flask/Django.
idk if you have answered this already but i guess you are changing the agent instruction instead of the description @29:23 so since you changed the instructions again later description is default.
Thank you for your the video
you really did a great job😊
Awesome vid Tim, We need more on AI agents ❤🎉
thank you very much... you solved my problem i was having with vector database for a week .. Thanks
Awesome tutorial. Gave me a lot of what I needed for my project.
tim is a genius thanks for all the helpful videos man
Thanks for the amazing content! It's super helpful and you make everything so easy to understand. Keep it up!
Great tutorial Tim, thanks for sharing. Could you tell me what's the most complex type of multi-agent workflow you've developed so far?
Thank you and congrats for this awesome tutorial !
I built this the other day with Node and React using langchain and supabase.
Could u share ur GitHub plz
4k stars were added because of this video!!
You’re a legend Tim! Thanks!
Such a good use case and video. Thanks!
All your videos are great, thank you
very inspiring as always, thanks a lot!
Great video, thanks Tim. But could you make additional video on how to evaluate nd analyse the performance as well as add bar charts
Thanks Tim Much appreciated
This is an excellent video. 💯
For those having problems with the AI not answering correctly or just saying I don't have information even though it is in the db whatever you are asking then do this. So for the Manager Agent controls, there is Agent Description, make sure to turn on the toggle so it's easier to edit and then write there:
A helpful assistant with access to the following tools:
1. FAQ
2. OrderLookupAgent
Replace with whatever names of the Agents you have, and everything should work fine. Do this for other Agents as well, if their description is empty, so just write something.
I tried your solution but facing same error
Yeah this does it !! Thanks Tim.
Thanks for creating such great tutorial. May I know how I can make my Langflow multi-agent as an API? Much appreciated.
Thanks man, this is awesome.
Getting a 'failed to load data' error when trying to uplaod the sample csv data @28:44.
i chatted with someone at datastax and it works now
Yea! But i think they have fixed that bug recently.
so no error's on my side.
@arhan_ai it decided to work now lol
Thanks a lot. Muito obrigado!!!
Great tutorial. Thank you for the detailed update from the previous langflow episode, since they updated so soon. I followed everything and got a slightly different test response from the frontend:
"I'm currently unable to provide specific details about the cancellation policy. However, generally, most companies allow cancellations within a certain timeframe after placing an order, typically before the order has been processed or shipped.
If you have a specific order in mind or need assistance with a cancellation, please let me know!"
I have all of the data setup in my database as followed in your video. Do you know why this happened and how it can be avoided?
thank u so much for this video.
been trying for some time and cant get it to show any product or order data - always reports it does not have access to it. Followed step by step a few times and same results. I wonder if something has changed in langflow that requires an additional step? Or am I missing something - others report the same thing in the comments
did you manage to solve the problem by any chance
i'm having this problem too, any updates?
I'm having this same problem too
the same problem
@@Mellybontest @theonlytapy @wakeboardr886 @Moths-m2r - ive not solved it yet no, kind of had to leave it for now as didnt have time @TechWithTim - Would be good if you have any ideas Tim? Great video as always BTW
Hello, please enable the option for Portuguese (Brazil) in your videos. The community here will be happy and eager to follow you. Thank you, and congratulations on the great video!
Hi Tim, great job on this! I need your advice on debugging the flow where the system accesses the "products" and "orders" tables. My flow is identical to yours but isn’t retrieving any data from the tables.
I noticed that disconnecting the FAQ module and ManagerAgent resolves the issue, and the "products" and "orders" responses work as expected. It seems the ManagerAgent has trouble handling results from both tools. The same issue occurs when I import your JSON file into LangFlow. Appreciate your help!
i am facing the same.
did you figure out?
@sachindhital4284 not yet!
the same
Did you guys figure it out yet?
I was facing a similar issue. In my case, it was because the field "Astra DB Application Token" in Astra DB tools for look ups was incomplete. I don`t know why it didn´t select ASTRA_DB_APPLICATION_TOKEN by default, but anyway, once I completed this it started to work.
I'm getting this error could you please help
ValueError
Details: Error performing search in AstraDBVectorStore: Ordering on non-clustering column query_vector_value requires the column to be indexed
This +1
same here
any fix ?
I encountered the same issue. In my case, my PDF initially consisted of plain text without proper titles, or visually structured paragraphs. To address this, I later added titles for each paragraph, organizing the content into questions, as shown in the video, and my issue was solved. I wish you are able to resolve that problem too.
Any one find a fix? I am encountering the same problem
Awesome 💯👌🏻
Tim, @38:06 your token accidentally leaks.
I get a 'failed to load' message when trying to load the sample data into the collections. Any ideas why by any chance?
any fix ?
@@1spotlightseries not yet, hopefully someone who knows a solution will reply
I'm also getting this error.
I was able to fix my error, but unsure if its the same as yours. I went to customer overview, then on the right hand side I generated a new token, and updated my astradb token key. It seems to work now. It is also possible it was just on the datastax side.
when i try 34:49
i get some information that does not exist in the csv file, on top of that, i get some columns that don't even exist int the csv file like this : [Shipping Method: Standard Ground
Shipping Address: [Your Address]]
i'm using llama3-70b-instruct.
is this a problem with the model ? because i tried a lor of nvidia models and they either say i'm a conversational model i can't do this, or i recieve an error saying the model doesn't support what i'm trying to do
Can you please make video on TTS model specially Voice cloning ones? they are incredibly hard to make so we all can use your help.
Thankyou = "brother"
print(Thankyou)
print ("we love u from india!")
Great video, thanks a lot !!!
Is it possible to use a csv instead of a pdf?
I would like to download a csv containing a table from a database to obtain answers from the questions asked (chat-to-csv)?
Hi Tim! I am getting UAM errors in the agent components. Can you tell me how to deal with it?
Hi Tim - Can we connect to a portal , login with user name and password using Agent & query?
When i run ManagerAgent at that time i got an Error which is :-
Error Building Component
Network error. Please check the connection to the server.
The "DS Astra DB" does not show "search input" filed in my case. Did something change in langflow DS Astra DB component after you released the video?
I created a database and loaded the data but for some reason the agent is nit able to fetch the orderID or prooductID or anything related to it
when I tried making a db, but it shows that my provider and region parameters are not available and not being abled to make a db
, what should I do?
Sir, I have tried using the Ollama model and need to connect it with the agent to access the tools. However, I am unable to do so. Is there any possible way to connect Ollama with the agent? Thank you.
Excellent ..!
Can i add Guided Projects from Coursera on my Resume and portfolio or do you suggest i build my own Projects from scratch?
Hi! I have a doubt related to the flow of the AI app. Is it really necessary to drag and drop the Vector Database again when loading the PDF file? I mean, I passed them to Ingest Data of the existing Vector database, and it seems to work. I'm not sure if I'm missunderstanding something.
Thanks a lot :)
Can you guide me about fully fledged freelancing
How can we deploy all this to our server or cloudstation or Streamlit cloud?
Hello I'm big fan
I am facing some problem using "pip" please make a video on pip problem's.
I'm trying to build the same thing, but instead of OpenAI, to use Mistral AI - but it does not work - is it enough for Agent to select Other language moel (not OpenAI) and link Agents Language Model pink dot to Mistral AI Language model pink dot - or does it need some other connection as well from LLM to Agent?
I used Google Gemini. I added the Google Generative AI under Models. I made my Agent Language Model is custom. Then connected the model to agent by Language Model
can it automatically get orders from my site?
Can u please use only free model and all?
i dont know when somone is following all the exact steps like you do and still getting errors how is it possible like i feel all that you did was false, like someone cant get the result even though doing same steps i am frustrated now on some stupid error damn
How is this agentic Ai different from Generrative AI, In gen AI we have aa asimilar project we can make a chatbot which can give ans to multiple pdfs at same time. Isn't this project similar to the one taught in this video? Plese explain anyone
Great
Cool 🆒
Python full course please
Bro Code on YT.
I'm getting this error after trying to run the multi-agent in the playground:
Error Building Component
Network error. Please check the connection to the server
Can anyone help?
SAME ISSUE, found a fix?
Would be better if you would create from scratch using langchain and deploy it using mlflow
Can I intergrate it in whats app and create a whatsapp bot
You all forgot giving star on github
Hello Tim your token leaked at 38:08
I mean... you did that on purpose right?
This is literally the best way to pay to win. You buy an extremely good computer, and your enemies will be stuck with 10fps on a normal machine (I guess the fps drop is also on the machine of the enemies
It is support language Arabic
Ai basic course
First pin pls
First comment
Wasn’t this the guy who was sleeping on AI 😂🫵
No bro he wasn't
you were sleeping on his channel
@DIY_Foodie mmm didn't see this coming
He made videos on AI like four years ago lmaooooo what the hell are you talking about
@@riderjohnny5117 He is a talking about you 🫵😂