Flowise AI Tutorial #1 - Introduction & Setup

hello and welcome back in this video we will have  a look at flow wise flowwise is an open source   tool that makes it easy to prototype Lang chain  applications using an easy to use user interface   in our launching JS series we had a look at how  we can build our own AI applications using Lang   chain and node.js but in this series we will  have a look at how we can create AI applications   without any coding experience flowwise is free to  use and can run locally on your own machine it can   also be deployed to a web server for online access  flowwise allows us to create AI applications using   a very simple drag and drop user interface and  can be a useful tool for prototyping Lang chain   applications before writing a single line of code  we can also use flow wise to expose API endpoints   for interacting with our flowwise flows directly  from our applications in this video we we will   have a look at how we can set up flow wise to run  locally on our own machines and we'll also have   a look at some of the examples that come bundled  with flow wise so let's talk about what you need   in order to follow along with this video first  you do not need any coding experience to follow   along with this video you will need node.js to be  installed on your machine so go over to nodejs.org   and download and install the LTS version of  node.js then open up the command prompt and   enter the following command npm install hyphen  G for Global installation and then flow wise go   ahead and run this command this will go ahead and  install flowwise on your machine while flowwise   is installing I do want to mention that there are  alternative ways of installing flow wise and you   can see these options by going to the flowwise  GitHub repo and then scrolling down to the quick   start section for this day demo we are simply  installing flowwise globally on our machine   but there are alternatives if you want to use  Docker there is an option for that and under the   developer section we can set up our own project  folder and then simply clone the repository and   install the dependencies but for this demo we'll  simply stick to the global installation once   installation is complete we can start up flow  wise by typing npx flowwise start if everything   was set up and installed correctly you should see  a message like this saying that flowwise server   is listening at Port 3000 so in order to access  flowwise we can now go to the address bar and   type in localhost 3000 and you will now see this  dashboard for flow wise from this dashboard we now   have a few options we can now create a new chat  Flow by clicking on add new and this will give   us this blank canvas where we can create our AI  application to the left of the screen we've got   this button for adding nodes and after clicking  this button we can see all the components that are   available in line chain like agents chains chat  models document loaders and more we can also use   the search nodes input box to find components for  instance when I type in open AI I can see the chat   models offered by openai as well as some other  openai components on the right hand side of the   screen we have a button to save our chat flow  as well as a settings button that will change   depending on the content within our flow we also  have this chat button which we can use to interact   with our application we can enlarge this pop-up by  clicking on this expand chat button we can clear   the chat history with a clear chat history button  and we can close this chat window by clicking on   this button let's go back to the main dashboard  from the main dashboard we can show and hide   these items by clicking on this burger menu button  on the right hand side we can change between dark   mode and light mode with this toggle next to  that we have a settings button where we can load   existing chat flows or export our current chat  flows within the menu we've also got Marketplace   as well as API Keys we'll get back to Marketplace  in a second but let's first chat about API Keys   one very cool feature of flow wise is that we can  expose our models or our chat flows to external   applications using our API keys and we will cover  this in this tutorial series if we have a look at   Marketplace we can see pre-built examples of chat  flows created within flow wise and there are quite   a few examples here as an example let's look at  the translator example so we can click on this   template and it will show us the chat flow for  this AI application please don't worry if these   components do not make sense to you yet as we will  be going into detail on what these components are   and how to use them in our project but basically  what we have here is an instance of our openai   model and just below that we have a definition of  a prompt template The Prompt template is used to   Prime the AI model with the system message and  in this example we are telling the AI that it   is a helpful assistant that translates some text  from an input language to an output language for   the human message we are simply grabbing the  input or the text that is parsed within this   chat window as input into the model and for the  variable list we are defining the input language   as English and the output language as Fringe we  are then using a chain to link the openai model   to The Prompt template and that will give us a  result if we want to go ahead and my changes to   this Marketplace model we can simply click on use  template and this will create a new chat flow for   us so the first thing we need to do is to save  this jet flow and give it a name something like   translation chatbot and I'll save this in order  to use the openai models we need to provide an   open AI API key in order to get an openai API key  you need to go to platform.openai.com and create   a free account once you've logged in you need  to generate an openai API key you can do this   by clicking on personal then click on view API  Keys then click on create new secret key give it   a name like flowwise Ai and then click on create  secret key you can now go ahead and copy this key   and store it in a safe place I will be deleting  this key after the recording so please ensure that   you use your own key for this we can then go back  to our chat flow and paste in the API key in this   field optionally we can change the openai model  that we want to use by default it's set to GPT   3.5 turbo but if you do have access to gpd4 you  can change this to gpd4 instead but I will stick   to GPT 3.5 turbo as it is a much cheaper model to  use than gpt4 what we can also do is to change the   temperature on this model and this needs to be a  value between 0 and 1 0 simply means that the AI   will be factual and will not be creative where a  value of 1 means that the model has free reign to   be as creative as it wants to be I'll just set  the value to something like 0.7 to give the AI   plenty of room to be creative but still factual  we can also click on additional parameters to   change other attributes like changing the maximum  tokens that we are willing to spend for each API   call for the prompt template I will simply use  the values that they provided but basically the   system message is studying the AI that it needs to  translate some text from an input language to an   output language the curly brackets indicate that  the input language and the output language are   variables that we will set down here so input  language in this string will be replaced with   English and the output language will be replaced  with fringe for the human message the value that   we enter in the chat box will replace the variable  of input again because of the curly braces this   becomes a variable field before we can test this  we need to save our changes by clicking on Save   I've fallen for this trick so many times you just  won't believe if you do not save your changes the   chat will not work correctly so if you run into  any issues or inconsistencies please ensure that   you've saved your chat flow this passing I love  programming and let's send this message and after   a second we get the response back from openai  which is now translated I love programming into   Fringe if we wanted to change the output language  we can simply change it in our prompt variables   for example let's change the output language to  Afrikaans if we run this chat now and let's pause   in I love programming again you will notice that  we are getting Fringe back as the output language   and that is because we haven't saved our changes  let's go ahead and save these changes let's run   the chat again passing in I love programming and  this time we are getting the response back in   Afrikaans so like I said earlier please ensure  that you save your changes before testing the   chat we now have our first AI application up and  running what we can do now is click on settings   and if we wanted to we can delete this chat  flow if we no longer want it we can create a   copy of this chat flow and we can also export  this flow if we wanted to expose this model   to the outside world or some third-party  applications we can simply click on API   endpoint and this will give us the code to embed  this chatbot using a JavaScript import or we can   see the python code as well as the JavaScript  logic and we also have a curl example for no   code Frameworks like webflow if we go back to the  dashboard we can see all of our created chat flows   right over here as an added bonus I'm going to  show you how you can quickly start up a flowwise   server and open it in the browser automatically  typically when you want to use flowwise you have   to open up the command prompt and then type  npx flowwise start note because we've already   installed flowwise it's not necessary to run the  install flowwise command again all we have to do   is execute npx flowwise start this will now  start up the development server but then we   still have to go to the browser and then open  up localized three thousand so what we can do   instead in order to run everything automatically  is we can open up notepad and then in this file   you can just paste in the following text and  you can find this text in the description of   this video then we need to save this file and  you can go to your desktop then for the file   name type in something like flowwise but because I  already have a file with this name I'll just call   it flowwise temp for now dot bat and as save  as type ensure that you've selected all files   go ahead and save this file after saving this you  can now simply double click on this file and this   will automatically open the command prompt the  start flow wise and it will automatically open   up your flowwise session in the browser in the  next video we will have a look at creating our   very first chat flows from scratch thank you for  following along if you enjoy this content please   consider subscribing to my Channel and let me know  in the comments if there are specific components   in flow wise that you would like me to cover in  a future video I'll see you in the next one bye

Channel: Leon van Zyl

Views: 17,849

Keywords: langchain, langchain tutorial, machine learning, flowise ai, flowise ai tutorial, flowise tutorial, flowise setup, flowise install, data science tips, artificial intelligence, ai, tutorial, how to, llm, openai, gpt4, flowise, chatbot, pdf, document, artificial intelligence course, pinecone, vector database, embeddings, openai embdeddings, langchain tools, langchain agents, langchain memory, ai chatbot, chatbot tutorial, chat with data, ai for business, chatgpt for company data

Id: tD6fwQyUIJE

Length: 12min 10sec (730 seconds)

Published: Wed Jun 07 2023