Build POWERFUL LLM Bots EASILY with Your Own Data - Embedchain - Langchain 2.0? (Tutorial)

[Music] have you ever wanted to easily create large language model powered Bots over any data set well you can now with embed chain this is a framework in which you can easily create any large language model powered bot over any data set that you want such as doc files you have Json Files YouTube videos PDFs site Maps notion and much more now embed chain abstracts the entire process of loading a data set chunk set and creates embeddings and stores it in the vector database for example if you wanted to create a chat bot for a certain type of video or a certain type of topic from a blog post all you need to do is input the links that are associated with that blog post or whatever data set that you're trying to upload and what embed chain does is that it takes care of creating the whole data set from the links that you have provided now this is an easy approach in creating powerful Bots to deploy for personal help has lost for commercial use and much more now throughout today's video I'll be showcasing how you can create powerful large language model powered Bots as well as going more in depth on embed chains so that you can get a more detailed analysis of what this application is all about so with that thought let's get right into the video now to install this application of embed chain all you need to do is install pip install embed chain now there's two ways to access this application you can obviously install it locally as well as accessing it off the cloud now this is something that they have uploaded on Google cloud and I'll showcase this as well as you go further onto the video but at the start I'm going to be just showcasing how you can install it locally on your desktop now first things first you're going to need to have git and this is an application that will help you clone the repository onto your desktop secondly you'll need Python and thirdly you'll need Visual Studio code you're also going to need an open AI API key for the whole application to function as it'll be creating the chat bot through open ai's API key so once you have all these things ready we'll get to the next step as we go on to the repository if you guys would like to access our private Discord definitely do so by checking out the patreon link in the description below now if you guys haven't followed World debate I I highly recommend that you do so so stay up to date with the latest AI Trends and lastly make sure you subscribe to notification Bell like this video and check out our previous videos as there's a lot of content that you will definitely benefit from so with that thought let's get right into the video the first step is to clone the repository and what you need to do is go onto the GitHub repository and click on this green button over here and copy the link for this repository now leave all the links in the description below so you can access it fairly easily now what you want to do first is open up command prompt if you have Windows now once you are here you can type in git clone and paste the link and click enter now I'll start cloning the repository now once this is done you easily just need to type in CD embed chain once you're here you just need to legit just copy this code over here and put it into command prompt and click enter and it will start unpacking the dependencies for the application and it will install the requirements that are needed for embed chain and that's easy as that you're able to install this application just for with two or three clicks now once this is done you can just open up visual studio code and input the open AI API key now guys once it has finished installing you minimize this or even close it and open up visual studio code now once you have opened this up you need to create a new window open the folder of what you just cloned so I gotta find the main chain and select this folder and open it up now what you need to do first is go on to dot m dot example and this is where you're going to paste your open AI API key so I'll be right back and paste my key now I have inputted my key and what you need to do next is just rename this file take out the example so you just need to take out the example and the dot click enter and it will be activated inside the system and now you just need to click save and you're easily able to run the application now guys they have uploaded the embed chain like documentation in which you can deploy the code of what chatbot you want with this simple code over here and you can basically just create your own python file in which you can input the things that you want so copy and paste this and put it into Visual Studio code into a new file and you're able to start creating your chatbot off the code over here where you just need to input your API key as well as the embedded online resources that you want to upload for creating the chat bot now they have also talked about different types of things such as inference types adding the data if it's not a certain type of link you can upload it through uh Digital Data and I mean not digital but like you can upload the data you're able to see the different sets of datas and these are some of the things that we'll talk about a little while later into the video now that we have showcased how you can install it locally we're also going to Showcase how you can play around with it on Google collab so with that thought let's get to the next step of the video where I'll showcase this now first things first what you want to do is go click on file and save a copy in your drive now once this is done you want to change the runtime change the runtime type to the best GPU that you have if it's the best hardware that you have select that one and once you have done that you can click save now what you want to do first is click on pip install and this is where it's going to start installing the dependencies for this application now this is for the people who want to run it on Google collab it's fairly easy and while this is going on what you need to do next is import the dependencies and you just need to click on this button over here now once this is finished installing I'll be right back and I'll install this other one so we can get to the next step where we just need to input our API key now once it has finished installing the dependencies you just need to Simply put your API key over here in which you can just click paste it and click run now once this is done uh you need to move on to the next step and this is where you add different types of data sources using embed chains.ad uh in the brackets and you basically just need to input The Links of the different online sources that you have now if you guys are uploading different types of content you can definitely do so with Google Drive in which you can link The Links of the Google Drive data sets that you want to upload and you can paste it over here now in this case we're making a chat or a chat bot that is responding to different types of things that are related to Elon Musk and we can see that there's different data sources or data sets that is retaining the data from Wikipedia Tesla which talks a little bit more about Elon Musk you have a YouTube video and this is a video I believe that is a podcast so the Elon Musk so in this case it has gotten three different types of data sets that are associated with Elon Musk and from this that you just upload these links and the chat bot is being created from embedded chain now this is something that can easily generate different types of like capabilities for you to answer and question the chat bot to about Elon Musk and you can see over here you give it a query how many companies does Elon Musk run and we can see an output that Elon Musk runs multiple companies some of the notable ones include SpaceX Tesla neurolink and the boring company and we can see it just gives you a answer based off the data that you gave it so this is an easy way for you to utilize this Tech to create any sort of chatbot that you want and it's fairly cheap and it's easy to deploy at any moment that you want so that's easy as that you can upload I showcase how you can install it locally as well as showcasing how you can access it on Google collab and that's basically it for showcasing how you can utilize in that chain now let's get to the next step of the video where we talk a little bit more about creating these different sizes chat Bots and going more in depth about this project so with embed chain there's actually different types of app types and there's actually three of them so we can see the first one is the data set app and this is in which it allows you to work with various data types such as like the doc files we each talked about Json Files YouTube videos PDFs site Maps notion and much more so from this you can upload or link these types of data sets to the framework and what embed chain will do is that will take care of the segmentation of the data into manageable chunks and it generates the embeddings that are needed to store them into the vector database so this type of app allows you for preparing and organizing the raw data to fuel your new language model chat bot now this is the first type of app the second type is the bot app and this is something that focuses on creating and deploying the actual language model powered Bots now this is by empowering embeddings that are stored in the vector database so that you can easily build these chat Bots that you can interact and respond to curies with and this is something that will provide assistance based off the content that you have linked and to the vector database now it's something that you can use for many different things customer service personal use uh embed it into your own workflow and this is something that can streamline the process of turning your data into functional conversational agents now thirdly you have the dashboard app and this is something that serves as a control center for managing and monitoring your language model Bots and this app basically offers a user-friendly interface and it enables you to track the performance of your Bot as it's going to be analyzing how you utilize the patterns as well as adjustments that are needed whenever you're conversing with the chat bot and it is able to actually provide you the interactions the user feedback as well as metrics that Empower you to optimize the chatbot's effectiveness and these are basically the three types of apps that are so considered with the embed chain now one thing you want to know is that these apps use open as models and they're all paid models so this is not something that is free in a way where you need to pay for the open AI API key so this something that requires you to pay and have a connected billing account if you're going to be utilizing these apps now under documentation they have done a great job in basically deploying this embed chain into different workflows as well as in different applications now they've integrated with a full stack they've integrated it with different API servers Discord Bots slack Bots telegram Bots and WhatsApp app bots so if you are looking towards deploying it on these different types of applications you can definitely do so by clicking on each and every one of them and it gives you a detailed analysis as to how you can deploy your own chatbot for these different types of apps another cool thing is they have a Community Showcase section in which it gives you a detailed uh templates as well as blog posts as to what people have done to create their own chat bots so this is a great way for you to build on like get the ideas that you want to build upon your own chatbots now they also have different types of things such as query configuration as well as custom configuration uh blog posts in which they give you the best types of things that are needed to create the best chat Bots and there's just more detailed analysis as to what you can do with this but in a way this is like Lang chain 2.0 you're able to create your own baby like chain with this project as you're able to deploy these chat Bots from embed chain fairly easily from a few clicks and it's easy as that guys and I hope you got some sort of value throughout today's video and that you're able to create your own chat Bots that are powered from powerful large language models so thank you guys so much for watching I hope you enjoyed today's video make sure you check out the patreon page if you guys want to access our private Discord if you want to check out our Twitter page definitely do so so you can stay up to date with the latest AI Trends lastly make sure you subscribe turn notification Bell like this video and check out our previous videos as you'll be able to get a lot of content a lot of value that will definitely benefit you so with that thought guys have an amazing day spread positivity and I'll see you guys fairly shortly peace

Channel: WorldofAI

Views: 8,353

Keywords: machine learning, artificial intelligence, langchain tutorial, embedchain, build llms, build llm bots, flowise, chainlit, embed chain, langchain, langchain agent, Build LLM Bots EASILY, langchain demo, Embedchain, chatbot framework, AI-driven chatbots, language model-powered applications, model training, bot deployment, AI technology, Embedchain benefits, AI innovation, chatbot technology, AI chatbots, chatbot creation, language models, AI framework, chatbot development

Id: jE24Y_GasE8

Length: 12min 23sec (743 seconds)

Published: Mon Aug 28 2023