LLAMA-2 🦙: EASIET WAY To FINE-TUNE ON YOUR DATA 🙌

Подписаться 163 тыс.

Просмотров 164 тыс.

50% 1

In this video, I will show you the easiest way to fine-tune the Llama-2 model on your own data using the auto train-advanced package from HuggingFace.
Steps to follow:
---installation of packages:
!pip install autotrain-advanced
!pip install huggingface_hub
!autotrain setup --update-torch (optional - needed for Google Colab)
---- HuggingFace credentials:
from huggingface_hub import notebook_login
notebook_login()
--- single line command!
!autotrain llm --train --project_name your_project_name --model TinyPixel/Llama-2-7B-bf16-sharded --data_path your_data_set --use_peft --use_int4 --learning_rate 2e-4 --train_batch_size 2 --num_train_epochs 3 --trainer sft --model_max_length 2048 --push_to_hub --repo_id your_repo_id -
🦾 Discord: / discord
☕ Buy me a Coffee: ko-fi.com/promptengineering
|🔴 Patreon: / promptengineering
💼Consulting: calendly.com/engineerprompt/c...
📧 Business Contact: engineerprompt@gmail.com
Become Member: tinyurl.com/y5h28s6h
💻 Pre-configured localGPT VM: bit.ly/localGPT (use Code: PromptEngineering for 50% off).
⏱️ Timestamps
Intro: [00:00]
Auto-train & installation: [00:17]
Fine-tuning - One Liner: [02:00]
Data Set Format: [05:30]
Training settings: [08:26]
LINKS:
autotrain: huggingface.co/autotrain
autotrain GitHub: github.com/huggingface/autotr...
All Interesting Videos:
Everything LangChain: • LangChain
Everything LLM: • Large Language Models
Everything Midjourney: • MidJourney Tutorials
AI Image Generation: • AI Image Generation Tu...
#llama #finetune #llama2 #artificialintelligence #tutorial #stepbystep #llm #largelanguagemodels #largelanguagemodel

Наука

Опубликовано:

17 июл 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 227

@samcavalera9489 11 месяцев назад

Thanks SO MUCH brother! You are a true hero! Fine tuning is the most important part of OS llms. That's where the value/wealth is hidden. I cannot wait for your following fine-tuning video.🙏🙏

@jersainpasaran1931 11 месяцев назад

Thank you very much champion! We are getting to the true spirit of open source, allowing science to be truly scalable for the public and public interests.

@OpenAITutor 11 месяцев назад

So great! Thank you for being so clear!!! loving it

@PickleYard 11 месяцев назад

Wow, just what I needed. I just put together a Flan Orca style dataset, I cant wait to try in Colab! Thank you for your hard work.

@engineerprompt 11 месяцев назад

Nice, good luck

@arjunv7055 11 месяцев назад

One of the best video I have come across. I will definitely share this channel with my colleagues and friends who wants to learn more on this topic.

@engineerprompt 11 месяцев назад

Thank you!

@LainRacing 11 месяцев назад

Very disappointed you didn't show this actually doing anything. How to verify or test if its working. I can run a script and have it do nothing... How do we see it actually worked or test it.

@bardaiart 11 месяцев назад

Thank you very much! Looking forward to the dataset preparation video :)

@garyhuntress6871 11 месяцев назад

I was initially skeptical but this was an excellent short tutorial. Thanks!

@engineerprompt 11 месяцев назад

Glad it was helpful!

@teleprint-me 11 месяцев назад

I was in the hospital because my lung collapsed and I've been having a seriously rough go at it lately (life long issues with fam, etc), so I really appreciate this video. Thanks for all your hard work. Researching these topics and understanding them is no small feat. Keep it up.

@engineerprompt 11 месяцев назад

I am really sorry to hear that! Hope you are recovering well. Wishing you a quick recovery. Also really appreciate all your contributions. Stay strong my friend!

@immortalsun 8 месяцев назад

Hope you get better!

@lallaaichakone2106 11 месяцев назад

wooow, after days of seraching for videos. I see everything that i wanted in this video and in simple terms. Great work

@engineerprompt 11 месяцев назад

Happy to hear that!

@miriamramstudio3982 11 месяцев назад

Thanks for the update. Very interesting.

@christianmboula8923 3 месяца назад

Superb tutorial by its clarity, simplicity and to the point...big Thank you! NOTE Bugfix : replace the underscore with corresponding dash to make the autotrain command run on colab

@adriantang5811 11 месяцев назад

Great Sharing again. Many thanks!

@bahramboutorabi5971 11 месяцев назад

Great video. Thank you

@sohailhosseini2266 9 месяцев назад

Thanks for sharing!

@sb98052 9 месяцев назад

Thank you for these very clear videos. Do you have any thoughts or pointers on resources for doing this type of training on code models such as CodeLlama?

@learn2know79 11 месяцев назад

Hi Thanks for the detail explanation. Could you please make another video explaining the RLHF with code implementation.

@karthigeyan88 11 месяцев назад

Hi, thanks for the video, could you explain in detail how to load the model and create an inference api in the local machine? that would be really helpful. thanks in advance

@MuhammadFhadli 11 месяцев назад

hi, have you find a way to do the inference?

@karthigeyan88 11 месяцев назад

@@MuhammadFhadli yeah, we have provisioned a Nvidia 64GB GPU machine and created an inference pipeline with llama.cpp library. Using an GGML model versiom from TheBloke huggingface

@immortalsun 8 месяцев назад

‘Could you explain in detail […]’ Talking to him like he’s ChatGPT

@serenditymuse 11 месяцев назад

The major work looks to be in making your dataset properly. Which is pretty common. Do you have or are you planning another video that is for training models simply by handing it a lot of files of say web content or better still the raw urls and perhaps something like tags and such? In other words how to add to unsupervised learning from a corpus.

@dec13666 8 месяцев назад

Nice video! A recurring aspect I have seen amongst these tutorials however, is that they never mention how to use the custom LLM model (i.e., doing some inference with the custom LLM model), or how to obtain metrics about it... Do you have any other video, where you discuss those 2 topics? Thank you!

@bagamanocnon 11 месяцев назад

how can i incorporate my own data into the 'assistant' fine tune? for example, a 100 page document about a company product. do i format it into the something similar to what's in the openassistant dataset and add it to the dataset? or finetuning on own data will be another finetuning step? i.e. after finetuning on the openassistant dataset, i need to run another finetune for my own data? cheers and thanks for all your hardwork to share your knowledge to us!

@ilyaskydyraliev6498 9 месяцев назад

Thank you for the video! May I ask, how big of a dataset should I have to see that fine tuning actually worked and model learnt new data?

@photojeremy 11 месяцев назад

would be great to have a colab notebook for this that included inference on the finished pushed model

@MuhammadFhadli 11 месяцев назад

hi, have you find a way to do the inference?

@manujmalik9843 11 месяцев назад

@@MuhammadFhadli did you find it?

@gerardorosiles8918 10 месяцев назад

I was thinking that once you push to huggingface you could use something like text generarion webui to play with the model

@AA-rd6nm 11 месяцев назад

Very deatiled thanks for sharing. I ❤ it.

@engineerprompt 11 месяцев назад

You are so welcome!

@zhirongchen9861 11 месяцев назад

Hi, how can I choose a method to finetune the model. For example, if I want to use LoRA to finetune lamma2, how can I do it?

@MaralSheikhzadeh 7 месяцев назад

well explained video. thank you:)

@engineerprompt 7 месяцев назад

Thank you

@swauce507 10 месяцев назад

After you finetune the model, how do you use it as a chat interface to query the model and see its results?

@user-hz3oh3xc9t 11 месяцев назад

Can you make a video on fine tuning a llm model on a recipe dataset.

@krishnareddy9 11 месяцев назад

Thank you for the video, I am looking forward video about how to prepare our own dataset without using huggingface dataset !!

@engineerprompt 11 месяцев назад

It's up now, enjoy!

@bookaffeinated 2 месяца назад

@@engineerprompt video link please.... And this one-line command throws error on colab: unknown argument, any suggestions pls?

@Koyaanisqatsi2000 10 месяцев назад

Thank you very much! Where can I view the loss of my training or evaluation data using this method?

@waelmashal7594 11 месяцев назад

Just amazing

@engineerprompt 7 месяцев назад

Thank you!

@Yash-mk8tc 11 месяцев назад

how to use this trained model? can you please make video on this?

@user-me5zg7is3k 8 месяцев назад

Great video thank you! I have a question; I have a prompt, an output from a model, and a desired output, how I can format this data, please?

@dr.aravindacvnmamit3770 4 месяца назад

Hi, the way you are explaining is very positive !!!! One solution am not getting is If I want to train my custom data on regional languages how to proceed can you share your knowledge on this. Which model is best on this and if we pass the Prompt in English will it gets converted to regional language and generates the ouput?

@sharadpatel107 11 месяцев назад

can you please put in a link for a colab notebook for this

@anjakuzev592 11 месяцев назад

Please make a video for creating your own dataset and actually using the model

@engineerprompt 11 месяцев назад

That is work in progress.

@vijayendrasdm 10 месяцев назад

What is the relation between max token size and the model kind of repeats itself ? The one you talk in the things to consider

@prakhargurha267 11 месяцев назад

2 questions. Is autotrain-advanced fine tuning is only available as a CLI format, or any other technique i available?Do we need collab pro for llama-2-7b-bf16.Can you suggest some smaller models to try?

@jongheebae6269 6 месяцев назад

I have the autotrain error as follows. autotrain [] llm: error: the following arguments are required: - -project-name So I changed '--project-name' instead of '--project_name'. Then faced another error.

@deepakkrishna837 8 месяцев назад

Hi Great Video. Thanks a lot for this. QQ: if I am building an information extractor and the max token length of the training data is 2750 and hence I have kept model_max_length as 3000. Do I need to strictly keep the block_size as well to 3000? Please answer!

@noraalzamil2660 9 месяцев назад

Thank you very much 🙏 Can I apply it with TheBlock llama-2-7b ggml?

@ShiftKoncepts 11 месяцев назад

I am a little confused, so the Llama LLM on gpt4all has to be trained first before usage with local docs?

@MarceloLimaXP 11 месяцев назад

Thanks guy ;)

@PickaxeAI 10 месяцев назад

What GPU should we select to complete this training? Could the T4 handle it?

@VerdonTrigance 4 месяца назад

How to train on unstructured data (a book for example) with self-supervized train algorythm and eventually make a chat from it?

@emrahe468 11 месяцев назад

finished running the autotrain in about 6h. And upload the model to hugginface. so what to do next? How to use this?

@nitingoswami1959 11 месяцев назад

Can we train this model on any data or it requires some specific format ? Does every llm requires some specific tabular data or any raw data ?

@alx8439 11 месяцев назад

Does it use lora or qlora techniques?

@user-nj7ry9dl3y 11 месяцев назад

For fine-tuning of the large language models (llama-2-13b-chat), what should be the format(.text/.json/.csv) and structure (like should be an excel or docs file or prompt and response or instruction and output) of the training dataset? And also how to prepare or organise the tabular dataset for training purpose?

@rainchengcode4fun 10 месяцев назад

timdettmers/openassistant-guanaco has introduction about the dataset, it should be a list of json with instruction, response in it.

@GEfromNJ 8 месяцев назад

See this is one the thing that gets completely glossed over in videos like this. If you take a look at timdettmers/openassistant-guanaco, you'll see that it's some nicely formatted data. It doesn't answer the question about how someone would take their own data and get it into this format.

@mdfarhananis8950 11 месяцев назад

Please teach how to create dataset for finetuning

@oxydol3456 2 месяца назад

learnt a lot from the video.Thanks. Is it easy to revert the model to the state before a tuning?

@engineerprompt 2 месяца назад

Thanks, yes, you are merging the extra "LoRA Adapters" layers to the model. The actual model actually remains unchanged so you can just reuse it for other purposes.

@Noscov 9 месяцев назад

Thanks for the video. I have a further question. At 5:50 your dataset has the columns instruction and input. What is the input-column for?

@immortalsun 8 месяцев назад

For example a question.

@justabacteria 11 месяцев назад

Could you explain or make a video on how to use your new fine-tuned model?

@engineerprompt 11 месяцев назад

Yes, that's coming very soon

@okopyl 9 месяцев назад

Amazing, but how to do the inference properly with this peft thing?

@georgekokkinakis7288 11 месяцев назад

I really love your tutorials, they are deeply informative. I was wondering for the following. Unfortunately 😔 all these LLMs are trained in English , but the world has so many other languages. If I follow the fine tuning you described in your video would I be able to fine tune the lama model for a specific dataset which has questions about mathematical definitions and methodologies with their according responses written in Greek? The amound off samples is about 100 questions with answers, I know it is really small but could this give good results for thebspecific dataset? And one last question , do you know any multilingual LLM which supports Greek. Thanks once more and keep up with your excellent ❤ presentations.

@AymanEL-BACHA 10 месяцев назад

hi @georgekokkinakis7288, have you tried training with your 100 sample/questions ? any improvements ?

@georgekokkinakis7288 9 месяцев назад

@@AymanEL-BACHA No I haven't yet

@bfam7110 11 месяцев назад

Is there embeddings or RAG with this approach?

@fangxiaoyuan-fm6vr 11 месяцев назад

Could you introduce how to deploy our model to a website? Thanks！

@ilhemwalker9145 4 месяца назад

hey please i copied the same line but i'm getting error : autotrain [] llm: error: the following arguments are required: --project-name. i don't know what to do

@contractorwolf 11 месяцев назад

subscribed!

@engineerprompt 11 месяцев назад

Thanks :)

@youwang9156 7 месяцев назад

thank you for ur video, literally save my life, just have one little question about the prompt format, you were using ### human and ### Assistant, so does this format basically depend on the pre-train model prompt format? like Llama-2 chat which has a certain unique format, but some like the Llama 2 base model, if there's no specific mention of that, then we can define our own format for the prompt? do I understand it correctly ? Thank you for your video again !!!!

@engineerprompt 7 месяцев назад

Glad you found it helpful. The template depends on whether you are using the base or the chat version. For the base model, you can define your own template as I am doing here because there is no template for it for using it as assistant (base model is actually the next word prediction model). But if you are finetuning a chat version then you will have to use the specific template that was used for finetuning the model. Hope this helps

@caiyu538 6 месяцев назад

How to save the fine tuned model to local disk instead of pushing to hub. Could you show us the model pushed to hub? These video graphs will make it clearer. Great.

@ajlahade2201 10 месяцев назад

can you please make a video on how to push this model to hugging face (like production level with model card) and call that model

@adapalarajyalakshmi3728 11 месяцев назад

Thanqu for the video can u explain how to use postgress database dataset

@Dave-nz5jf 11 месяцев назад

you would probably need to pull the data in batches, in the right format, and then run this autotrainer on a batch basis. But it's an interesting question - if you have data that's changed (in the database), and you retrain the model, how does the updated data impact the model output.

@arjunv7055 11 месяцев назад

some of my friends who followed this tutorial mentioned they see an argument issue. I think it is because of the command being broken down into multiple lines. Running the command in multiple lines requires a '\' to be added at the end of every line. Final command should look like this !autotrain llm --train --project_name '' \ --model TinyPixel/Llama-2-7B-bf16-sharded \ --data_path timdettmers/openassistant-guanaco \ --text_column text \ --use_peft \ --use_int4 \ --learning_rate 2e-4 \ --train_batch_size 2 \ --num_train_epochs 3 \ --trainer sft \ --model_max_length 2048 \ --push_to_hub \ --repo_id /'t \ --block_size 2048 > training.log &

@nayyershahzad8051 6 месяцев назад

getting following error, kindly help: autotrain [] llm: error: the following arguments are required: --project-name

@user-wy3rr4dp2c 6 месяцев назад

Hello, I am a beginner in LLM. I generated the model folder locally according to the video operation, but the folder size is only about 130Mb. The base model I use is 7b llama2. Is this normal? Why is the model size reduced so much? How do I get the normal size model? I would be grateful if you could answer it for me

@sravanavvaru4473 11 месяцев назад

hey the thing I did not get is on what data is the model getting trained ??

@SadeghShahmohammadi 11 месяцев назад

It took a few hours, everything went well but at the end the model is not in my hf repository! Cannot find it anywhere!

@nufh 9 месяцев назад

Other than google colab, what is other platform that we can use? I'm still new, just started to learn about python.

@sanj3189 11 месяцев назад

How can i use LLama2 for generating synthetic data

@machineUnlearner 4 месяца назад

i have a time series data, with 7 to 10 parameters. What should I do ?

@MicaleAntonio 7 месяцев назад

Does auto train do multi-label text classification?

@pareak 5 месяцев назад

What is the difference between the SFT and the Generic trainer?

@tubesarkilar 8 месяцев назад

can you show a sample of time series data file to feed into Autotrain?

@anantkabra6825 8 месяцев назад

Hello I am getting this error can someone please help me out with it: ValueError: Batch does not contain any data (`None`). At the end of all iterable data available before expected stop iteration.

@DikHi-fk1ol 8 месяцев назад

Please make another tutorial on how to fine-tune a model on custom dataset rather than using the hugging face ones.

@Shahawir 11 месяцев назад

I wonder if it is possible train LLAMA, on data where input are numbers and categorical variables(string), of fixed length, to predict a timer series of fixed size, anyone knows if this possible?

@user-zt1ie5ir2p 11 месяцев назад

I haven't tried it on colab yet but was wondering, do we need colab pro or colab pro+ for this tutorial?

@engineerprompt 11 месяцев назад

For this, you can use the sharded model with free version but for full model you will need pro

@8eck 11 месяцев назад

What if i only want to feed a specific non-instruction data into the model? For example some financial data or some books or some glossary? Can i just keep the ###Output empty, will the model learn from that data? Also, do i need to split that data into train and test parts or it is not required and is optional for pre-trained models?

@curtisho5255 11 месяцев назад

i have the exact same question! omg!

@phoenixfire6559 11 месяцев назад

If you leave the output empty then the model will learn to give you empty responses every time you put that type of data in. The best way to make the data for your finetune is thing about it from reverse. When you put the input in, what do expect the output to be? That's what you should be filling output with.

@8eck 11 месяцев назад

@@phoenixfire6559 i'm talking about pre-training like fine-tuning, models in the pre-training phase doesn't get any output examples, they just learn from the data, that's what i'm trying to understand. Is fine-tuning is only about question & answer pairs? How to continue pre-training of the model with frozen base weights. Just like transfer learning.

@curtisho5255 11 месяцев назад

@@8eck exactly. he don't get it. We want it to train on pure data, not train on Q&A responses. He must have not played with chatbase.

@robosergTV 11 месяцев назад

@@curtisho5255 lmao the author of the video knows this. The video is clickbait for farm views (which is money) from noobs, who cant use simple google search.

@nexusinfosec 11 месяцев назад

Could you please create a video on the dataset creation?

@VadiyalaRR 10 месяцев назад

ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE--ui8YKz4d-E.html hope it helps you

@pickaxe-support 10 месяцев назад

Is there a link for the google colab notebook?

@Yash-mk8tc 11 месяцев назад

can you make a video on hugging face basics

@meteor1 8 месяцев назад

Can I fine-tune llama-13b-GPTQ using autotrain-advanced ?

@bharatkaushik9916 9 месяцев назад

Can someone tell how to inference this model ?after pushing it to hub thanks

@gamingisnotacrime6711 10 месяцев назад

I have a custom dataset with 50 rows. For how many epochs should i fine tune thr model? Each line in my dataset is in this format - ###Human: Who is John?### Assistant: John is a famous youtuber (My dataset has only a single column named text and 50 rows which have the data in above format So also are there any issues with my dataset?

@prestonmccauley5467 11 месяцев назад

I followed this exactly in collab, but seems that something is wrong with the arguments, Can you share your colab file?

@arjunv7055 11 месяцев назад

if you are breaking the command into multiple line please make sure to add \ towards the end so finally the command looks like this !autotrain llm --train --project_name '' \ --model TinyPixel/Llama-2-7B-bf16-sharded \ --data_path timdettmers/openassistant-guanaco \ --text_column text \ --use_peft \ --use_int4 \ --learning_rate 2e-4 \ --train_batch_size 2 \ --num_train_epochs 3 \ --trainer sft \ --model_max_length 2048 \ --push_to_hub \ --repo_id / \ --block_size 2048 > training.log &

@deepjyotibaishya7576 11 месяцев назад

Colab always stuck and show me complete on 57% when it running on merging It it possible to upload folder to Hugging face and laster on can i Mergin it and make it ai model ??

@ElNinjaZeros 10 месяцев назад

Thanks for sharing, by the way does auto-train need to be paid to be able to use it?

@engineerprompt 10 месяцев назад

The cli version is free to use unless they changed something recently

@rafaelferreiradesouza9972 11 месяцев назад

I'm looking for a way to create a local server, that uses my trained IA for answers like a personal assistent, can anyone tell where can I learn that?

@okopyl 9 месяцев назад

Now when i generate responses, i get input generated as well. Why? How to avoid that?

@hvbris_ 11 месяцев назад

Really cool video, saved mea lot of time, how much memory would the GPU need to train a model like llama2-13B - is 12G enough or should I consider getting something beefier? Thanks in advance!

@engineerprompt 11 месяцев назад

Atleast 16GB

@souvickdas5564 10 месяцев назад

Can we fine-tune LLaMA model on MNLI or SNLI dataset? Is it worth doing ? Give me your thought.

@engineerprompt 10 месяцев назад

Yes, I think it’s possible. These might already be in the training data.

@user-jx3wy6fe4s 5 месяцев назад

I am facing issues in the autrain line where its stating argument should be project-name instead of project_name and even if i change that its not taking arguments like data_path, use_peft. can someone help me out?

@jodter1 11 месяцев назад

in visual studi code??

@BatoolZ-q5r 21 день назад

do we have to add the tiny pixel model to colab?

@titangadget 4 месяца назад

I'm using this one line training code but is giving me error... can you update it?

@ScottzPlaylists 3 месяца назад

🤯 Wow Wow Wow ❗

@engineerprompt 3 месяца назад

thanks :)

@Noshiru 2 месяца назад

Hello! The question might be stupid, but how come this is so difficult to learn to the AI our own data ? I mean, when you talk to ChatGPT for example, if you tell it stuff, it will remember (if you use the same chat) what you said and it will be able to answer your questions about it. Why can we just give the AI a documentation for example ?

@user-bs5xo4nd1t 7 месяцев назад

how to create the own dataset from the pdfs

@Homboy_ 9 месяцев назад

Which llama 2 based model can you recommend for text classification problems?

@engineerprompt 9 месяцев назад

The bigger the better.

@Homboy_ 9 месяцев назад

@@engineerprompt ok thanks, and can't I tokenize my data and give to model for tuning? Also if when u give just 1 column text data as input and Target column is text classification like fraud/normal. What should be the input format in CSV