That stretch is a death trap. Hope your commute is never made shorter via ambulance, stay safe out there.

I would love to know how you prepare the ahi, that sounds incredible!

Username checks out hehe. Jk, don’t get why people are downvoting you for being a lil wrong!

I have mixed feelings about this. I am getting my bachelor’s in CS in May and have (ENTIRELY unexpectedly) gotten to the final round of interviews for an MLE job; I have done no formal course work on anything AI related, but I have had two internships in ML. I feel like I have sold myself short for the software engineer, data scientist, or even MLE I could have been with a bit more work by BS-ing my way into a field I barely understand, and I know if I do somehow get this job, I will be scrambling to learn everything I don’t know on the fly. Having bombed plenty of interviews so far, it’s become painfully clear that this might have been a mistake. If you’re going to do ML, dive in head first and try to deeply understand it - take any course you can on it and start reading papers NOW to best prepare yourself for getting a job when you graduate. It IS possible, but you also want to be good at what you do and not end up in way over your head. I know I’m headed for at least a master’s degree later even if I get this job because this field is evolving too quickly for me to grasp the major concepts and keep up with the current science. Sorry if this is rambling, these are just my thoughts being in the position I am currently, hopefully this is helpful in some way!

What is the difference between fine-tuning and transfer learning? Can you do deep learning without either of those things?

Is it true that learning rate matters less if you’re using an adaptive optimizer? If so, would you argue for or against using learning rate decay, and why?

Where did you get that ring?