This AI Research Agent Generates Detailed Report Using PhiData

Sdílet
Vložit
  • čas přidán 27. 04. 2024
  • 🌟 Welcome to an exciting tutorial where we dive into the world of AI research with cutting-edge tools like Llama 3 and Phi Data! In this video, I'll guide you through setting up a powerful AI research agent capable of processing data at an astonishing rate. From adding URLs for auto-research to uploading files for detailed reports, I cover it all! 🌐
    🔍 What You'll Learn:
    Setting up Phi Data and Llama 3 for rapid AI research.
    Configuring your system to read PDFs, data from URLs, and search the web.
    Generating in-depth reports with references using the latest AI models
    🛠️ Setup Steps:
    Clone and setup Phi Data from GitHub.
    Install necessary libraries and activate environments.
    Integrate Grok and T API for enhanced data processing capabilities.
    💡 Make sure to subscribe and hit the bell icon to not miss any updates on AI and tech tutorials. Like and share this video to help others in the tech community! 👍
    🔗 Resources:
    Patreon: / mervinpraison
    Ko-fi: ko-fi.com/mervinpraison
    Discord: / discord
    Twitter / X : / mervinpraison
    Code: github.com/phidatahq/phidata/...
    Timestamps:
    0:08 - Introduction to Phi Data and Llama 3
    0:34 - Step-by-Step Installation Guide
    1:54 - Setting Up Virtual Environments
    2:02 - Installing and Configuring APIs
    2:53 - Launching the User Interface
    3:06 - Demonstrating URL and PDF Data Processing
    4:59 - Advanced Report Generation Techniques
    5:36 - Customising the Code and System
  • Jak na to + styl

Komentáře • 33

  • @maheenbaber7515
    @maheenbaber7515 Před měsícem +3

    Hello Marvin, could you please guide how to build a full fledged ai maths and stats solver that can answer after reading text or scanning files and images, display equations in latex and sketch graphs as well. I actually want to embed this solver as a SAAS on my website. Curiously waiting for your expert guidance.

  • @RobynLeSueur
    @RobynLeSueur Před měsícem +2

    Nice, simple and fast. Will give this a go, thanks.

  • @codelucky
    @codelucky Před měsícem +2

    Is it possible to solely utilize the GROQ API without the need for Ollama embedding or local resources? This will enable me to upload it to a smaller server and have access to it via a phone browser, which would be great.

  • @sfl1986
    @sfl1986 Před měsícem +2

    Please explain how to modify the structure (system prompt?) of the article so that it uses a differenr template. Also how to chsbge the lenght of the final article (number of words)

  • @adamchan4403
    @adamchan4403 Před měsícem

    Thanks for sharing

  • @phidata
    @phidata Před měsícem +1

    We love you Mervin, you are awesome 🧡

  • @kate-pt2ny
    @kate-pt2ny Před měsícem

    Thanks, the mouse track is beautiful

  • @jackgaleras
    @jackgaleras Před měsícem

    Gracias .. lo voy a intentar ..

  • @jial.5245
    @jial.5245 Před měsícem

    Marvin, thank you so much for producing such cogent videos with an incredible pace. Been watching every one of your videos since subscribing.
    One question I have on phidata: is it fair to say phidata doesn’t support multi-agent framework similar to how autogen does?

    • @MervinPraison
      @MervinPraison  Před měsícem +1

      True, it doesn’t support multi agent framework yet

  • @GospelAIWorship
    @GospelAIWorship Před měsícem

    Amazing

  • @waqaskhan-uw3pf
    @waqaskhan-uw3pf Před měsícem +1

    Please make video about Romo AI- super AI tools in one place and learnex AI - world's First fully AI powered education platform. My favorite AI tools

  • @marioalejandromonroyg.6664
    @marioalejandromonroyg.6664 Před měsícem

    Hi Marvin, excellent tutorial, everything works fine, except that I can't run it, because in Windows it is not possible to use the export command, so the api keys cannot be set.

  • @stevey19861
    @stevey19861 Před 14 dny

    do I have to do ollama run nomic-embed-text?

  • @MeinDeutschkurs
    @MeinDeutschkurs Před měsícem

    My mind is exploding. Will I ever catch up? 😅 Great stuff. Next, I try to differentiate between several (local) vector databases, adding up sources. Also: deleting several sources if not needed anymore.

  • @alexalex4192
    @alexalex4192 Před měsícem

    ty so much, very helpful. Can you make a tutorial about RAG CHAT (the next option) ?

  • @user-gr8on2pc5y
    @user-gr8on2pc5y Před měsícem +1

    hi,
    Mervin Praison
    can you make a video of devin AI building an automated bot that books appointements

  • @ohardest
    @ohardest Před měsícem

    I got all the way to the "docker run -d \" part. command wouldn't work for me in Windows PowerShell.

    • @phidata
      @phidata Před měsícem +1

      The \ doesn't work in windows, Please can you run:
      docker run -d -e POSTGRES_DB=ai -e POSTGRES_USER=ai -e POSTGRES_PASSWORD=ai -e PGDATA=/var/lib/postgresql/data/pgdata -v pgvolume:/var/lib/postgresql/data -p 5532:5432 --name pgvector phidata/pgvector:16

  • @Armoredcody
    @Armoredcody Před měsícem

    Is there ways to ask follow-up questions without having to regenerate a report?

    • @MervinPraison
      @MervinPraison  Před měsícem

      You need to customise to perform this

    • @phidata
      @phidata Před měsícem

      @Armoredcody you can use the RAG chat instead of the RAG Researcher from the sidebar to chat without having to generate the report

  • @SamiSabirIdrissi
    @SamiSabirIdrissi Před měsícem +1

    First

  • @liuxingzhe
    @liuxingzhe Před 28 dny

    I got all the way TO "docker run -d \ " , however, it says "docker run" requires at least 1 argument.
    See 'docker run --help'.
    Usage: docker run [OPTIONS] IMAGE [COMMAND] [ARG...]
    Create and run a new container from an image