How to fine-tune a model using LoRA (step by step)

Поделиться
HTML-код
  • Опубликовано: 17 ноя 2024

Комментарии • 34

  • @solidkundi
    @solidkundi 4 месяца назад +3

    love it... as i am still a noob..would love to see a llm example with summarization model ..and to see the format involved .. thank you again!!!

    • @chiggly007
      @chiggly007 4 месяца назад

      Agreee this would be helpful to see

  • @flcor
    @flcor 4 месяца назад +1

    One note: the row-column decomposition is valid for matrices whose rows (columns) are not linearly independent - that’s probably why they train on the row-columns themselves and not on general matrices that cannot be factorized into row-column form. So, there’s clearly a tradeoff here between memory and linear independence.

  • @kaloyanmirchev5006
    @kaloyanmirchev5006 4 месяца назад

    Thank you for presenting such great ideas. My imagination surely goes wild when I attempt to think of possible applications..

  • @tecbrain
    @tecbrain 4 месяца назад +1

    Fantástico vídeo. Gracias por el tiempo que inviertes. Ahora me queda entender bien el código.

  • @UtopIA-IA
    @UtopIA-IA 4 месяца назад

    OMG! that is so powerful, thank you, I am alone doing projects of this type and this will be very useful for me, thanks for sharing you knowledge.

  • @olivierdulac
    @olivierdulac 4 месяца назад +2

    Thank you, great content

  • @mabadolat
    @mabadolat 4 месяца назад

    This is great stuff, Santiago! I wish you had posted this video a few weeks ago. We just completed our final class project where we trained five different BertClassifier models on five different tasks. Our fine-tuning and inference code structure is very similar to yours. We definitely could have used this approach to use just the specialized adapters instead of the full BERT models.
    However, I have one question: I'm not clear whether the full model will ever be used during this process after we get fine-tuned adapters, or just the fine-tuned weight matrix for evaluation and inference?

    • @underfitted
      @underfitted  4 месяца назад +1

      You need to use both: the general model + the finer tuned adapter. The adapter describes how the general model should change on the fly

  • @abdullah_mufti
    @abdullah_mufti 4 месяца назад

    very well explained

  • @alextiger548
    @alextiger548 4 месяца назад

    Fantastic stuff ,Thank you!

  • @restrollar8548
    @restrollar8548 4 месяца назад

    Awesome as usual!

  • @jayng4028
    @jayng4028 4 месяца назад +1

    Nice tutorial, would like to ask how to fine tune an AI model that generate interior design?

  • @monugarg8627
    @monugarg8627 3 месяца назад

    Hey sir. Very good explanation. Sir is it possible for you to make a video on Ai Agents and tools please.

  • @Leo-ph7ow
    @Leo-ph7ow 4 месяца назад

    Thanks!

  • @syedasadzaman9591
    @syedasadzaman9591 3 месяца назад

    Good Stuff! Santaigo.
    The channel name should be `Tutorials That do not Suck!` =}

  • @cyberlando
    @cyberlando 24 дня назад

    Can someone please explain to me how you would train an LLM adapter on tabular data that depends on the rows being consistent with each other ? I’m having issues with the llm pulling row 5s id and row 300s description

  • @devevangelista
    @devevangelista 4 месяца назад

    if run this with 16G memory and RTX 2060 could work?

  • @allanmogley
    @allanmogley 4 месяца назад

    How do you make those models that interact with data,
    Like I once saw someone create something really amazing that inteprets data from a a database and makes interpretations and reports from the data wothout hallucinating (It only fetches from the underlying DB)

  • @kencottrell
    @kencottrell 4 месяца назад

    When you have the original model + plus the Adapter model, can the original model still solve the save generic tasks? In other words, can you perform original inferencing tasks PLUS specific tasks?

  • @sirojiddinnuriyev2839
    @sirojiddinnuriyev2839 4 месяца назад +1

    There are a lot of contents how to fine-tune LLMs with LoRA or QLoRA. You gave us same food just with ‘apple genius’ keyword.

    • @underfitted
      @underfitted  4 месяца назад +1

      I’m glad you knew everything I said already! Good for you.

  • @krishnasoni8169
    @krishnasoni8169 3 месяца назад

    I cannot talk with the agent,the connection is established but it aint respond or neither taking image i/p please suggest something

  • @AngusLou
    @AngusLou 4 месяца назад

    Can you make a Google Colab notebook for the same fine-tuning?

    • @underfitted
      @underfitted  4 месяца назад +1

      Yes. Just load this notebook in Google Colab

  • @m.active
    @m.active 2 месяца назад

    there isn't a general solution to decompose a matrix of M*N into two vectors of M*1 and 1*N. If that was the case we could have some all the issues in the data compression by now. A lossless compression of 99.99% for huge matrix is a strange achievement.

  • @robertobreve8623
    @robertobreve8623 4 месяца назад

    The jupyter notebook has broken images.

  • @2021philyou
    @2021philyou 4 месяца назад

    Missing all the info that is needed to implement the idea on own data set

    • @underfitted
      @underfitted  4 месяца назад

      What info would that be?

    • @underfitted
      @underfitted  4 месяца назад

      What info would that be?

    • @avinashnair5064
      @avinashnair5064 Месяц назад

      @@underfitted can we train on our organisation data how can we do that

  • @Gaurishinnskills
    @Gaurishinnskills 6 дней назад

    Is It Free?