M2 Max VS M2 Air - Machine Learning. Should you buy the Air for Data Science?

Is the most powerful M2 Air good enough for serious machine learning? And how is it different from M2 Max.
Intel MacBook Pro will be the last one in our comparison.
#macbook #m2macbookair #m2max #machinelearning #datascience

[@John-qo9hw]

0:19 Light GBM, can you tell where I can find that code and that kind of dataset to test my mac. I have an old one and I want to know how much of an improvement it would be relatively. Thank you.

[@lifecrunch]

The dataset is proprietary and unfortunately I can't share it.
And the code is simple:
from lightgbm import LGBMClassifier
model = LGBMClassifier()
model.fit(X_train, y_train)

[@user-mj5nx8yf9b]

Thank you for this video! It is exactly what I am looking for!❤
I am debating between the macbook air and macbook pro. My only concern is the fan. Will the macbook air throttle when running traditional machine learning models, or when testing deep learning models with small datasets ?

[@lifecrunch]

For Deep Learning on small datasets and for just about any traditional machine learning task Air will perform just fine. And if you need to train a large neural network from time to time - you can use a free colab environment with a starter GPU.
Get the Air!

[@user-mj5nx8yf9b]

@@lifecrunch Thank you!

[@koushikroy8668]

The video is really helpful. But to confirm, do macbook M2 air 13 , do not have any GPU? I need to train and finetune llms, so asking. (free colab also have limited usage)

[@lifecrunch]

Hi. The M2 in the Air does have 10 GPU cores. But keep in mind, that you might need extra configuration of PyTorch or tensorflow to take advantage of the M2 GPU.
You can read this article on the PyTorch setup: reneelin2019.medium.com/mac-m1-m2-gpu-support-in-pytorch-a-step-forward-but-slower-than-conventional-nvidia-gpu-40be9293b898
And you can watch my other video on the tensorflow configuration: th-cam.com/video/QcMAHnHlIKk/w-d-xo.htmlsi=wwOGtcUIxpdRCPVN

[@harshraj9077]

Hi, I am confused between MacBook Air M2 and M3(8gb Vs 16GB) in both, which one to opt , undergrad student for Artificial Intelligence and Data Science course, please reply. Thanks

[@lifecrunch]

As for the processor (M2 vs M3) I would say either one is fine. But for ML tasks memory is crucial. Go with 16GB.

[@harshraj9077]

@@lifecrunch means for Data Science,AIML stuff M2,M3 is capable only need 16GB, okay one more question how is M1 for these stuffs?

[@lifecrunch]

I never used the M1 but I presume they are not that far apart. Either one will be great.

[@atharvmukane4768]

Will 16gb RAM in M2 Air be enough for machine learning?

[@lifecrunch]

It’s a good start. Plus MacBooks can swap RAM from SSD very efficiently.

[@geraldalanraj]

​Is the 16gb ram 1tb SSD M2 Air enough for Machine Learning and Deep Learning?