KZhead
Ғылым және технология

Group By Meaning...Not Keywords

Greg Kamradt (Data Indy)
2024 ж. 13 Мам.
2 023 Рет қаралды
Жүктеу

Stay up to date with Greg: mail.gregkamradt.com/
Semantic Deduplicator: github.com/gkamradt/SemanticD...
SingleStore: tiny.one/QUtq9Wa
Dive into the innovative world of semantic deduplication. Navigate the challenges of consolidating product feedback and explore the mechanics behind a new Python package I built. From refining grocery lists to streamlining KZhead comments, witness the transformative power of semantic understanding.
Outline
0:00 - Skit
0:26 - Introduction
0:53 - The Problem
1:15 - The Semantic Deduplicator
2:10 - How it works - Refactoring
2:36 - How it works - Multiple Requests
3:00 - Sponsor: SingleStore
4:00 - How it works - Deduplication
4:40 - Example: Grocery Lists
5:08 - Example: Feature Requests
6:00 - Example: KZhead Comments
6:49 - Get Started
7:09 - Ryan Brandt shout out
7:22 - Outro
Greg’s Info:
- Twitter: / gregkamradt
- Newsletter: mail.gregkamradt.com/
- Website: gregkamradt.com/
- LinkedIn: / gregkamradt
- Work with me: tiny.one/TEi2HhN
- Contact Me: Twitter DM, LinkedIn Message, or contact@dataindependent.com

Пікірлер

  • Great idea of something simple but very useful. Thank you.

    @RPhaF@RPhaF6 ай бұрын

  • Super helpful. Thanks Greg

    @shubhamtripathi1601@shubhamtripathi16016 ай бұрын

    • Nice - thanks Shubham!

      @DataIndependent@DataIndependent6 ай бұрын

  • quality video and quality content as always, thanks Greg!

    @jessaco.8653@jessaco.86536 ай бұрын

    • Awesome! Thank you Jessa Co.!

      @DataIndependent@DataIndependent6 ай бұрын

  • Thank you for your great tools.

    @caiyu538@caiyu5386 ай бұрын

  • thank you for the great video.

    @micbab-vg2mu@micbab-vg2mu6 ай бұрын

  • A cool follow up to this video would be showing how to use this within excel using their new built in Python features

    @janalgos@janalgos6 ай бұрын

    • Oo ya that would be cool - good call

      @DataIndependent@DataIndependent6 ай бұрын

  • Very cool. Would this work with non-English languages?

    @mads7869@mads78696 ай бұрын

  • 2:38 I don't get it, how do you solve the second problem - splitting multiple requests? Do you just use few-shot learning or embedding vectors?

    @borisrusev9474@borisrusev94742 ай бұрын

    • Yep exactly few shot learning

      @DataIndependent@DataIndependent2 ай бұрын

    • Was looking for this as well, thank you :)

      @Falstad88@Falstad882 ай бұрын

    • Great, thanks! Awesome channel btw, don't be surprised if I have further questions. I just discovered it and now I'm bing watching it :)@@DataIndependent

      @borisrusev9474@borisrusev94742 ай бұрын

  • Any way to solve this issue for big data like pyspark? Might wanna avoid LMs there

    @pushkaraggrawal@pushkaraggrawal6 ай бұрын

    • Hm...I'm sure there is but I'm not too familiar with PySpark to suggest a method

      @DataIndependent@DataIndependent6 ай бұрын

  • BTW Thanks ❤

    @camcodex@camcodex6 ай бұрын

    • Thanks Cam

      @DataIndependent@DataIndependent6 ай бұрын
KZhead