That's not the way to look at the numbers. First, you'd want to talk about whether the results are statistically significant. Second, when dealing with a fatal disease, people are pretty happy if their odds of survival go up by a few %.

Presumably because they're otherwise very healthy.

They are rarely at peak healthy. They are just in peak physical shape.

Depending on the sport - strain on muscles, joints, heart

You should stop presenting your opinions such as “Cancer only affects people who generally already have other problems” and “they will probably get something else anyway” as facts.

Imagine a reader who is not one of your lucky “most” majority. Imagine a reader whose cancer was not caused by the bad lifestyle decisions that you listed. Put yourself in the position of somebody who undergoes extensive surgery/radiation/chemotherapy and then lives with the side effects of these treatments. Consider what it’s like to live with the fear of recurrence even after such treatment. Then maybe you’ll understand why people might be excited about the potential of this sort of screening.

VPC Endpoint is a feature of VPC: https://docs.aws.amazon.com/vpc/latest/privatelink/endpoint-...

Disclosure: I work at AWS.

Here are some examples from my team’s 2019 work: We contributed numerous changes to containerd. We open sourced firecracker-containerd, and we also created a Go SDK that others are using to work with Firecracker. We contributed to Debian and the Debian kernel team. We contributed to Envoy. We collaborated with a number of communities, including Kata Containers, Red Hat’s Clair, and the Open Container Initiative. All of these examples are sustained investments, not one offs.

Sure, it's contributed where it's convenient. I still don't see a list of committers so that I can look at what those contributions were.

Yes. An example can be found in the issue about making this simpler to set up: https://github.com/aws/aws-app-mesh-roadmap/issues/2.

To clarify, an AWS customer has a shared responsibility to describe the security of their systems including how they use AWS tools, and in this respect Amazon is no different than other AWS customers.

Question for the author: how do you actually deploy your model? Do you have a dependency on Spark in your production system?

Since its just a dot product between the learned weights and the feature vector, we do this in the application layer as lacksconfidence surmised.

Being that this is a SVM, which is typically evaluated as a simple linear sum of weights, I imagine they reimplemented that in the application layer. Would be curious how they handled the normalization steps (reimplement that as well?)

Yep. We normalize our features as part of training, and the stdevs of each feature are part of the resulting model, along with the weights. (The means are always 0 because of the way we construct our training set.) The weights we use in production are actually normalized_weight / stdev.

I'm not 100% sure what you mean by utility real-time programs, but yes, this probably is they meant. Nest knows where you live (I don't mean that in a creepy way), and they can help you enroll in what are called "demand response" or "load control" programs in your area. When you enroll, you get an incentive like a bill credit or a free thermostat, and you give the utility permission to turn your AC off for a little while at some point during the summer.

Many of these utility programs actively recruit customers, so chances are (a) you'll get a chance to sign up even if you own a different type of connected (e.g. WiFi) thermostat, and (b) the sign up process won't be hard.

Some utilities (like ComEd in Chicago) have a real-time program where the price of electricity changes every 5 minutes in response to supply and demand.

https://rrtp.comed.com/live-prices/

I was asking if Nest can observe and respond to that data, which it looks like it still cannot.