> I repeat the above recursively on transitive dependencies as many times as necessary. I also repeat the cursory code review any time I upgrade a dependency.
If this guy has to work on a "modern" frontend project, he's gonna review dependencies until the heat death of the universe.
I'm not a huge fan of the format of the article, but I am a huge fan of a solar plant that has sufficient storage (and mindset) to provide power when the sun isn't shining. The more like this, the better, imo.
I live on this mountain. These are not settlements. These are logging camps or seasonal work camps. My grandfather worked up there and he was jealous of the Japanese for their beautiful shacks when the Norwegians were in canvas tents. Of course, he admitted that the Japanese just worked harder to make their own quarters.
There were orchards, livestock, farms of potatoes, onions and legumes, but most of the activity was in hauling out doug fir to tidewater with oxen or steam donkeys or building miles-long flumes to shoot WRC shinglebolts down to the river or Burrard Inlet.
Thanks for taking efforts to explain this — more people should be aware of this interesting effect.
I agree with all you have said except for the conclusion part. BN without L2 is not underfitting. In my experience it is overfitting due to small effective learning rate. It is easy to verify — just compare the gap between training and test losses / errors.
So my conclusion from the derivation is, L2 penalty in BN acts as a regularizer in a different way — by increasing the effective learning rate of the weights.
On a related note, this effect could present even without normalization. By adding just a scalar multiplier parameter in the branch, the weight’s scale could be more or less decoupled from its direction. For reference, I will make some shameless self-promotion here about our recent work on training residual networks without normalization:
http://arxiv.org/abs/1901.09321
Are we supposed to be sympathetic to the “problems” these people are facing selling their overpriced real estate? My only regret is that they’re not getting screwed over harder.
Doesn’t matter, nobody will use it. Taxi service in japan is good enough to where this won’t be a big factor. Ride hailing services are much more expensive than taxi’s.
I have not read this book but based on the article, I feel the concepts in Malcolm Gladwell's Outliers > This book. Relationships do not prove causality, but David tries to with the stories of Van Gogh and Tiger Woods and Federer? The examples in the article leave me frustrated because professional artists and sports icons are not a repeatable success. Using those people as examples of success is bad data. I don't comment, this article bugged me for some reason. My takeaways are: Specialization matters, how you find 'your thing' is up to you. And figure it out well before you turn 40, unless you are Van Gogh.
App companies claim that the "data it collects for clients is kept private and not sold" If you believe that, I can get you a great price for the scrap iron from the Eiffel Tower.