Easily Build Prompt Tuning & Prefix Tuning for LLMs: Soft Prompt Engineering Beats Fine Tuning

Рет қаралды 3,726

Күн бұрын

Пікірлер: 74

@maryammiradi 5 ай бұрын

🚀 Ready to Master Real-World AI and Land High-impact Roles in just 4 weeks? Join my Hands-on Training and Transform your Career Today 🎯: www.maryammiradi.com/training

@JoramMillenaar 8 күн бұрын

Hey Dr. Maryam! Thank you for taking the time to share your insights. I think you might have missed an important part of the prompt tuning process; training. You mentioned the random state of the embedding, which is initially true, but the power comes in when you train that embedding.

@maryammiradi 8 күн бұрын

Thanks alot! You are absolutely right I didn’t focus on Training in this video. It was more to show how prompt tunning and prefix are different than fine tunning and inspire people to try it.

@AdarshKumar-sj5dn 5 ай бұрын

I truly appreciate your hard work and your honest attempt to share your knowledge. 🙏

@maryammiradi 5 ай бұрын

@@AdarshKumar-sj5dn Thank you 🙏

@alishafii9141 5 ай бұрын

I appreciate you sharing your experiences and I have the utmost respect for your work

@maryammiradi 5 ай бұрын

@@alishafii9141 Thanks alot and glad it is useful 👋

@mnava4290 5 ай бұрын

Excellent tutorials Dr Miradi👏

@maryammiradi 5 ай бұрын

Thanks alot and I am glad it was useful 😄

@arielwertlen6709 3 ай бұрын

This is excellent, thank you Dr.

@maryammiradi 3 ай бұрын

@@arielwertlen6709 very glad to hear!!

@ommanojsharma7118 5 ай бұрын

Hello mam your unique projects series was great please add more unique project modules in it

@maryammiradi 5 ай бұрын

Sure, let me know which topics / techniques are most challenging to you and let me know if you are a data scientist

@ommanojsharma7118 5 ай бұрын

@@maryammiradi thanks for the response mam this no mam I am not a data scientist i am undergrad still in the college learning about the field of data science and this are the topics which i find challenging - 1) Doing the eda and feature engineering on the raw dataset and this is the most crucial step before implementing the machine learning solution 2) Not having the idea about how to implement the time series data end to end production ready system using the MLops tools 3) how to do the hyperparameter tuning of our deep learning architecture as when I do that it gets overfitted 4) how to implement the research paper using the pytorch as it is a most valuable and essential skill to become a top tier data sciencetist like you It would be very kind of you if you could tutorials on this topics Again thanks for your warm gesture

@pradeepjungkarki4510 5 ай бұрын

hello mam iam from nepal and has been loving your content.Iam expecting a lang graph series as well as other data science projects from you also free data science guide mam

@maryammiradi 5 ай бұрын

@@pradeepjungkarki4510 Thanks and Langgraph is already on my list. Here is the link to My Free Data Science Guide: www.maryammiradi.com/free-guide

@shettysajith7466 5 ай бұрын

Well explained .. thanks for the gr8 video

@maryammiradi 5 ай бұрын

Very glad to hear! If there are specific subjects in LLMs you want to know about let me know.

@sitimariyah5972 5 ай бұрын

Thank you so much for your amazing videos. This is really helpful for beginner in using Hugging Face.

@maryammiradi 5 ай бұрын

Really glad to hear it is useful to you 👋

@hendoitechnologies 5 ай бұрын

kindly post full course video about "Claude 3.5 sonnet model" fine tuning and API

@maryammiradi 5 ай бұрын

Will put it on my list. Thank you.

@MrDrumballer 5 ай бұрын

This is incredible , i also want Free Data science Guide

@maryammiradi 5 ай бұрын

Glad to hear! Here you go, here you can find the guide: www.maryammiradi.com/free-guide

@saaj-l9n 19 күн бұрын

Free data science guide

@maryammiradi 18 күн бұрын

Of course! Please find your free guide here: www.maryammiradi.com/free-guide

@authentic-reflections 2 күн бұрын

Free data science guide

@KumR 5 ай бұрын

Hello.. Is many shot prompting similar to fine tuning?

@maryammiradi 5 ай бұрын

@@KumR no, there are two different things. With many shot prompting you improve your the communication with your LLM to improve the outcome. With fine tuning you adjust the internal parameters of LLM to improve the outcome for example for an specialized task like medical or healthcare. Both improve the outcome but with different approaches.

@KumR 5 ай бұрын

@@maryammiradi Thanks mam

@michelleming6465 4 ай бұрын

Thanks for sharing! where can we find the code for prompt tuning and prefix tuning as you shown in your video?

@maryammiradi 4 ай бұрын

Of course! I Will put it a notebook and share it soon.

@michelleming6465 4 ай бұрын

@@maryammiradi thank you so much for your prompt response! It's very helpful!!!

@michelleming6465 4 ай бұрын

@@maryammiradi Hi Dr. Miradi, where should we find the notebook?

@maryammiradi 4 ай бұрын

@@michelleming6465 I am traveling right now. Monday I can put it in Notebook

@michelleming6465 4 ай бұрын

@@maryammiradi

@dhanugadu 4 ай бұрын

Free data science guide

@maryammiradi 4 ай бұрын

Of course! You can grab it here: www.maryammiradi.com/free-guide

@mustaphaachir6546 2 ай бұрын

Free data science guide

@maryammiradi 2 ай бұрын

Sure! Here you can grab your free guide: www.maryammiradi.com/free-guide

@865oct 5 ай бұрын

Free Data Science Guide

@maryammiradi 5 ай бұрын

Sure! Here you can grab the free data science guide: www.maryammiradi.com/free-guide

@techievans Ай бұрын

Free Data science Guide

@maryammiradi Ай бұрын

Of course! Please go ahead and grab it here: www.maryammiradi.com/free-guide

@mnava4290 5 ай бұрын

Free data science guide

@maryammiradi 5 ай бұрын

Of course! Here you go: www.maryammiradi.com/free-guide

@alishafii9141 5 ай бұрын

Free Data Science Guide

@maryammiradi 5 ай бұрын

@@alishafii9141 Here you can find y Free Data Science Guide: www.maryammiradi.com/free-guide

@srikanthdongala6922 Ай бұрын

Free Data Science Guide

@maryammiradi Ай бұрын

Of course! Go ahead and grab it here: www.maryammiradi.com/free-guide and you can explore my website here: www.maryammiradi.com

@bunmisadiq3298 5 ай бұрын

Free Data Science Guide

@maryammiradi 5 ай бұрын

Sure! Please go ahead and grab the Free Data Science guide here: www.maryammiradi.com/free-guide

@alexSehdevaTube 3 ай бұрын

Free Data Science Guide

@maryammiradi 3 ай бұрын

Of course!! Please go ahead and grab it here: www.maryammiradi.com/free-guide

@itsshaikaslam 5 ай бұрын

Free Data science Guide

@maryammiradi 5 ай бұрын

Of course! Go ahead and grab it here: www.maryammiradi.com/free-guide

@kayteeflick 4 ай бұрын

I really enjoy the content, but there are so many transitions/moving text or images in the slides that it becomes distracting and a lot of flashing. It was hard to watch the video.

@maryammiradi 4 ай бұрын

Thanks alot for the feedback!! I tried to make it as clear as possible but apparently it works the other way around and became distracting. Video Editing takes lots of hours and if it is not needed for you guys. If it helps you to keep it simple, It will only save me time.

@ommanojsharma7118 5 ай бұрын

@maryammiradi thanks f or the response mam this no mam I am not a data scientist i am undergrad still in the college learning about the field of data science and this are the topics which i find challenging - 1) Doing the eda and feature engineering on the raw dataset and this is the most crucial step before implementing the machine learning solution 2) Not having the idea about how to implement the time series data end to end production ready system using the MLops tools 3) how to do the hyperparameter tuning of our deep learning architecture as when I do that it gets overfitted 4) how to implement the research paper using the pytorch as it is a most valuable and essential skill to become a top tier data sciencetist like you It would be very kind of you if you could tutorials on this topics

@maryammiradi 5 ай бұрын

Thanks for all the suggestions. I will add them to my list.