sir what if i dont have an acc of google cloud biling enabled, so i cant deploy it to google cloud run?

are you running models on GPU or ANE?

hey really great video, you are super clear explaining complex concepts, thanks!

is it possible to set a custom url? or if its not possible on google cloud itself, is there some way to do so while still having google cloud have your container?

Why the x86 version of rust?

Saved me a lot of time. Thank you so much!

Thanks Faizi for sharing the instructions! Following the steps I faced 2 issues but luckily could solve them: 1) I could not create the rosetta terminal. Executing "env /usr/bin/arch -x86_64 /bin/zsh --login" in a terminal solved this. (btw I'm using os version "Sonoma 14.3") 2) After having installing pytorch and tensorflow, the tests failed because module "chardet" could not be found. This could be solved by running "pip install chardet". After that all worked fine. Thanks again!

why is mac so costly if everything is difficult on it

now how do I use that ml_env I am able to activate it on terminal but I how do make it on my .py files, How do I add this ml_env onto my folders

Wow I did it!! had a few extra steps for me, ended up doing it in iTerm in order to duplicate the terminal to make it Rosseta and that gave me a few extra steps I think but chatGPT worked me around a few errors I had but I got it, 99.9%! Thank you!

is it free?

Hey, is cloud run a good option for streamlit though? Streamlit session management for example is in memory, so the session data will be lost every time the instance is shutdown given that's how cloud run or serverless work in general. How did you go about that?

Ha Ha, and that's why I left apple to the dark side.

I LOVE YOU DEARLY! seeing [{'label': 'POSITIVE', 'score': 0.9998656511306763}] was the BEST NEWS after two days of back-forth with my M1!! LOVE YOU!

Anyone following this guide, if you need to run a terminal shell in rosetta on Ventura, just open a terminal window and type 'arch -x86_64 zsh'

Thank you so much, this is so helpful. While following your steps I faced many issues and solved most of it. The only part I'm really struggling from is when I try to install tokenizers=0.11.1. The error is that tokenizers=0.11.1 is not available from current channels, and when I tried to download it in version 0.13.3 the transformers will not work :( . would be really great and will save my time, if you could advise me what to do to solve this. thank you.

thank you so much for taking the time to put this tutorial together. saved me a ton of headache!

No idea why this was in my feed. Don't even know huggingface

Thank you so much Faizi! This is so helpful! (:

Thank you! I highly recommend this tutorial.

You are a true hero!

Thanks for this. Unfortunately, I couldn't get past step 2. When I put in the command "conda create --name ml_env python=3.8" I get back "command not found: conda". I tried it a few times with the same result. I have no idea how to proceed. Any guidance would be greatly appreciated. I'm not a coder or programmer so this is all foreign to me. My goal is to get Dreembooth running on my MacBook Pro (M1 Max chip).

This is an amazing guide, thank you!

You are a star!!!!!!!! Thank you so so much!!!

hi could I check if this works on Jupiter notebook? im trying to do hugging face there but am unsure how to do it

Thanks for the video, I was also looking to add in google's Oauth to authorize my users through Institution gmail. But it failed as I was encountering version conflict of protobuf, Please do let me know if you have any reference for this.

Hi there, thank you for your video and helpful tips. If the video goes well; would love to see you run a tutorial on doing one of your fine-tunes to see how it goes.

thank you so much ! but how to get this work with jupyter notebook

Hey, I didn't know about these details. Big thank you. Really. Appreciate the time you took to record and post this.

You might want to change the head line to TUNING…

if you start a business or make money, you need to master a high income skill to be able to do it... based on my own experience. enjoyed watching your video... a fellow creator

Hey, great content. Is it possible to get into contact with you to discuss a possible partnership?

that mouse pointer scared the hell out of me

super helpful thank you.

I was also unaware of hyperparameters. Keep up the sharing of your process. Lots of newbies out here as well.

Could you make a video on how to create a dataset for n00bs. Would get a lot of hits, I’m trying to build a chatgpt based on ancient texts, to ask generic questions, would appreciate it, thanks, also where did you learn how to do the datasets, any good material I could review? Thanks again

Thank you so much, Faizi!

Thanks

Hi Faizi, how is it going? Nice video, how about we hop on a call to discuss you helping fine tune the model to suit our business case?

This is awesome thanks bro!

Hi, downloading rust using your method got me the wrong version : stable-aarch64-apple-darwin installed. Now when trying to uninstall using 'rustup self uninstall' i get this error: command not found: rustup. Anyone else encountered the same problem?

any update on the ci cd vid?

Thanks for the detailed walkthrough. I have carefully followed your step and ran successfully until the part of installing “tensorflow-deps” where it says “PackagesNotFoundError: The following packages are not available from current channels: tensorflow-deps”. It’s been days looking and trying for solution for this m1 machine. No luck yet. Hope you care to help 😊

ᑭᖇOᗰOᔕᗰ ☀️

Many thanks for your nice article on the medium and this helpful video. Enjoyed every second of it. When you have the availability and the resources, could you please film the 2nd part (linking to github, and using continuous deployment), Sir? Thanks again

I did the same process as you did. Deployed it without any errors but when I open the public url the app is stuck in "Please Wait..." screen. Did you encounter such issue?

I'm using AWS but I can't get the ci/cd aspect of docker to work, so a vid on google continuous deployment would be good.

what interpreter did you use after? I got it working in CL however I need it to get it working in VS code but u get a "None of PyTorch, TensorFlow >= 2.0, or Flax have been found. Models won't be available and only tokenizers, configuration and file/data utilities can be used." error

I had installed transformers on M1 few months ago without having to install rust compiler but today I happened to uninstall Miniconda & Miniforge and after reinstalling them, I tried reinstalling transformers too and stumbled upon an error prompting to install rust compiler. Tried installing rust however encountered another error: "could not amend shell profile: '/Users/username/.bash_profile': could not write rcfile file: '/Users/username/.bash_profile': Permission denied (os error 13)" Was able to install only using brew however the arch type is arm64 instead of x86error, so that didn't work either. Sat through the entire video and since you'd created an environment to install Tensorflow and PyTorch which I already had in my M1 couple of months ago, activated the env and ran the following command: "conda install -c huggingface transformers" and it's working now. Although I do remember installing transformers using "pip install transformers" before I messed up my system today. To my surprise, I didn't have to install rust compiler although I tried to but failed and just activating the env and running this line: "conda install -c huggingface transformers" did the job. But thank you so much..Was helpful although I didn't have to go through the pain of installing everything.

Dope video. Keep it up!