Lesson 5 In-Class Discussion

(Clayton Yochum) #22

I suspect he’ll cover that in the ML class; replacing a “traditional” ML model with a deep learning model shouldn’t change the ensembling step.

(yinterian) #23

See the forums where there are a couple of pointers to articles about this.

(Vikrant Behal) #24

No fastai. I feel handicapped! :frowning:

(asaia) #26

Sounds great! Thanks @yinterian

(Pete Condon) #28

Some of the predictions are exceeding five, would this model perform better if the values were capped at zero and five?

(Walter Payne) #29

How did you decide on the dimension of that matrix? Why 5 rows and not x?

(Gerardo Garcia) #30

Are those numbers embeddings?
The random numbers

(Vikrant Behal) #31

Guess, that’s like features!

(Gaurav Rele) #32

embedding for movies and users

(James Dietle) #33

What are the differences between features and embeddings? Different words same thing?

(yinterian) #34

embeddings are learned

(Vikrant Behal) #35

Those features which aren’t explicit! But we want NN to find them for us?

(Pramod) #36

features need not necessarily be embeddings. feature could be as simple as ‘is the date a state holiday?’. Embeddings are representations of these features. I think they are the answer to the question : “How can we represent BladeRunner2049 numerically?”


Responding to the question “do you retrain for a new user or new movie?”

At this year’s Grace Hopper many companies mentioned that they approached that problem with the way Jeremy said, i.e., you’d have a specific new user model or new movie model, then retrain over time.

The “what are you interested in?” during onboarding (like meetup.com’s) is a popular approach, but IMO not very friendly if your user is not of your core user group for example.

(Ken) #38

What does it mean to try different n_factor sizes? Do you train and evaluate using MSE and compare with different embedding sizes given a number of epochs?

(Pete Condon) #39

Yep, it’s a hyperparameter … so AFAIK no one has worked out a way of automatically picking the best number.

(yinterian) #40


(Ken) #41


(Brian Holland) #42

“…wait, what?..”

It’s so comforting that happens to you too @jeremy!

(asaia) #43

Any resource that helps explain a hyperparameter? Thanks in advance