Getting Started with Deep Learning on MacBook Pro M1

Hi all,

First post here, just getting started in deep learning and AI, appreciate any help with this.

I’m trying to get started with deep learning/machine learning/AI. I recently bought a jetson Nano 2GB (4GB is out of stock), and began working on the DLI course. Though the background information is really interesting (as is the AI podcast which I regularly listen to), I feel really let down by the jetson nano. I’ve struggled immensely to get it up and running (from not having a suitable wifi connection, slow internet and then really really slow processing times), i’m almost at the point of giving up on the getting started with Jetson Nano course. I’ve just done the emotions model but the system crashes (every time) when I try to evaluate it in real time. So i’m pretty close to giving up on the jetson nano until a 4GB version becomes available.

Having said that, I really don’t want to give up on deep learning and I think the tutorials are laid out in a really good way.

I’d like to enrol on the ‘Getting Started with Deep Learning’ course, but as it’s another $90 I wanted to check and see if i’d run into the same pitfalls as I have on my jetson (so hence this post). I am using a MacBook Pro M1 2020. Which is a really great system but using Arm64 architecture. I know that Tensor Flow is only written for x86 architecture however, I have managed to successfully install it through Rosetta 2 and into a virtual environment. I have run other packages such as mediapipe this way and it works really well.

As the MacBook Pro doesn’t use an Nvidia graphics card (I believe it’s an ‘integrated GPU’) I wasn’t sure if this would pose any initial barrier to deep learning.

If you could advise me on this it would be greatly appreciated.

Many thanks

Sorry to hear you were having problems using Nano, and also sorry that your post didn’t get any replies.
I have moved this topic to the nano sub-category, folks there would be in a better position to help and advise on issues using Nano.
As for using your laptop for the DLI course work, I don’t have direct experience but generally the DLI course leverage libraries and code which have been tested on NVIDIA GPUs, so you should get the best experience when you have access to an NVIDIA GPU, if you laptop supports external PCIe extensions so you could add desktop GPU cards - but thats an expensive route, but I am hoping your nano keeps you going. GPUs can also be accessed online on AWS and Azure.
Good luck with your AI training.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.