How deep on deep learning?

[Ben-Jamin-Griff]

Hi Daniel,

I was wondering how in depth you're planning on going into deep learning? I'm relatively familiar with ML but I'm starting to learn deep learning with Tensor Flow and I'm already getting lost with loss functions and optimizers.

Cheers,
Ben

[mrdbourke]

Yo Ben, This is code-based so we write TensorFlow code for plenty of loss functions + optimisers to *use* them, however, we don’t go through them in-depth. In fact, we use only 4-5 of each (e.g. 2-3 loss functions + 1-2 optimisers) throughout the whole course. If you’re after in-depth resources on different loss functions/optimizers, I’d check out the ML glossary: https://ml-cheatsheet.readthedocs.io/en/latest/optimizers.html In short: - loss functions = how wrong is your model (lower is better) - optimizers = how should your model change its underlying patterns to improve the loss score Daniel

…

On 18 Feb 2021, at 6:42 am, Benjamin Griffiths ***@***.***> wrote: Hi Daniel, I was wondering how in depth you're planning on going into deep learning? I'm relatively familiar with ML but I'm starting to learn deep learning with Tensor Flow and I'm already getting lost with loss functions and optimizers. Cheers, Ben — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AD7ZOCJNIHJLY6XMPZ5QHF3S7QS4BANCNFSM4XZBO2HA>.

[Ben-Jamin-Griff]

Hey Daniel,

Thanks for explaining this, that resource is super useful thanks!

The course sounds great, can't wait for the release.

Cheers