selljamhere

Tesla has been publicly traded for 10 years.

https://finance.yahoo.com/quote/TSLA/

I don’t see any arguments in those slides about trying to “even out black vs white ppl deaths metric.” Social inequality is one of several factors included in their ethical and logistical plan, but interpreting that as an implied black vs white issue seems like a very uncharitable view of the author’s intent.

+1 for Harvest. It saves me a lot of time generating hourly invoices.

https://www.getharvest.com/

Would it be possible to add a flag for the compiler to disable the line table pre-expansion?

The slide demonstrates a concrete implementation of the interface, showing how it can be used to communicate with the real file system. The mock implementation is left as an exercise for the reader, as you and I would likely have different requirements.

Initializing through explicit constructors e.g. `func NewSample(a, b int) Sample {...}` can also add a compile-time check. Of course the onus is once again on the developer to remember to use the `New` funcs instead of instantiating directly.

Edit: spelling

Does anyone know why it is only supported for a Model 3 B+?

H2-Flow!

I doubt the manager focused on the hours. It was a simple way to say the manager doubled the expected timeframes when estimating tasks. By assuming the team is productive 1/2 the time, the task should take 2x the initial estimate.

I’ve used UpCounsel for business documents and filings. They’re a pay-as-you-go service, and might have what you’re looking for.

https://www.upcounsel.com

Can someone summarize the changes between 1.13 and 2.0? Given the major version increment, I'd assume there's a major paradigm shift.

Is there anyone who can ELI5 what `beta8.Mod(beta8, curve.P)` does?

Awesome!

Does the move to gVisor pave the way to connecting CloudSQL instances? The removed limitations sound like they were road blocks.

I always think back to Ze Frank’s “True Facts”: https://youtu.be/gNqQL-1gZF8

I disagree. When used to call a human a dinosaur, it is an abstract term. People are fully aware that the referenced person is not actually a prehistoric reptilian. I think referring to all such prehistoric creatures as dinosaurs is problematic because it doesn’t allow for the same obvious distinction. How could one otherwise know if the article refers to all prehistoric reptiles or Dinosauria?

Like "upperware" with a "t".

I disagree. The article notes several times that a source requiring broad anonymity should be viewed with skepticism. The implication is that the source my have ulterior motives, and the journalist has a duty to determine the story’s veracity.

However, not all sources require anonimity simply to spread disinformation. Some of our most culturally significant news stories have spawned from sources requiring extreme anonimity. In these cases it is up to the journalist to corroborate the information.

TL;DR - Building your own rig is fun, but can be a pain to maintain. There is a path on Google Cloud to start ML projects for free, and scale up when you need to.

I think you're hitting the biggest up/down sides on the nose.

There is an innate joy to building your own rig (I'm guilty of it myself), and it does provide a fair amount of flexibility. The two main downsides that come to mind are the up front cost for parts, and the maintenance required to keep the system up to date. I've had success with a CentOS rig running nvidia-docker[1]. The annoyances mount over time though, as OS and Nvidia driver releases can be a pain to keep up to date, especially if you're working with several of the major frameworks. This is the main reason I've moved most of my ML projects to the cloud.

Cloud has many benefits including scalability and minimal maintenance. There are many options in the cloud ML space, but my current favorite leverages Google's cloud. I'll get into detail below, but the main reason is that they have several tools that build a natural progression and will grow with you as your projects mature.

First Steps:

I'd recommend checking out Colab[2][3] to get your feet wet. It includes a FREE GPU instance, and is a Jupyter based notebook, so you can easily utilize any of the python based frameworks. (1/2 of a K80, if memory serves. 12 GB) As with all free resources, there are some caveats. Colab is designed for interactive usage, so they don't encourage long running processes. However, the general consensus online is that you can get around 12 hours of training in at a time before the instance restarts itself, which is plenty of time to kick off a job and save checkpoints to cloud storage for resuming later. All in all, it's a great tool get a project off the ground. One bit of strangeness (from a software engineer's perspective): the notebooks are stored in google drive. This adds an interesting possibility for collaboration (it's in the name, right?), but leaves something to be wanted in terms of source control.

Growth:

If Colab isn't cutting it, there are a couple next steps. If you need more compute power, or have long running training jobs, the Colab UI can actually be run locally backed by your own GCE instance (with attached GPUs)[4]. If you need additional services (databases, map/reduce, etc) Datalab[5] is Colab's bigger beefier cousin, with all sorts of integrations across the Google Cloud Platform, as well as 3rd party providers.

EDIT: Link formatting

[1] https://github.com/NVIDIA/nvidia-docker

[2] https://colab.research.google.com/notebooks/welcome.ipynb#re...

[3] https://research.google.com/colaboratory/faq.html

[4] https://research.google.com/colaboratory/local-runtimes.html

[5] https://cloud.google.com/datalab/

One of my favorite examples of public speaking is Mr. Rogers testifying before a Senate subcommittee in support of funding for national public television. He is calm, thoughtful and thorough through the entire process, speaking simply and without grandeur. It's a small wonder he found such success, and a great example of the importance of clear communication.

https://www.youtube.com/watch?v=fKy7ljRr0AA

>> It's pulled, and it's fine, but there's clearly a balance between "octopus merges are fine" and "Christ, that's not an octopus, that's a Cthulhu merge".

This made me chuckle.

One of the best benefits I can think of favoring Core Data and iCloud/CloudKit is the very public stance Apple has taken on privacy.

One largest downsides I can think of is vendor lock-in. If you intend to expand beyond iOS, there will be some friction and a great deal of re-implementation once you select another platform to build on.

If you do intend to move on to other platforms, I'm impressed by Realm's offerings [1]. The support local data storage, as well as syncing to their cloud platform (with a monthly fee, of course).

EDIT: Forgot to include the [1] link.

[1] https://realm.io/

In perfect timing, Bank of America released Erica, "your virtual financial assistant."

https://promo.bankofamerica.com/erica/

That makes sense. I was thinking in terms of global lookup keys, which would require hashes to persist across runs. When thinking in terms of object equality, the hashes need only live as long as the objects themselves. Namely, for the lifetime of the program.

>>New hashing algorithm

>>Hash values vary from run to run (so don’t depend on hash values or order)

I must be reading this wrong. One of the requirements for a good hash function is determinism - specifically that any given value in the input space maps to exactly one value in the output space.

grep is the product of a lot of thought and engineering. There's a great rundown by GNU grep's author, if you're interested [1].

One big takeaway is that grep doesn't simply traverse the string linearly, instead using the Boyer-Moore algorithm [2].

1: https://lists.freebsd.org/pipermail/freebsd-current/2010-Aug...

2: https://en.wikipedia.org/wiki/Boyer%E2%80%93Moore_string_sea...

EDIT: Interestingly, there seems to be an internal Boyer-Moore implementation in Golang's strings package[3], though it only appears to be used in strings.Replace("")[4]

3: https://golang.org/src/strings/search.go

4: https://golang.org/src/strings/replace.go?h=stringFinder

This seems like a natural progression toward integrating Tensorflow with Apple's CoreML. With Tensorflow Lite's CoreML conversion, a native Swift frontend could open doors for ML on iOS.

> artificial intelligence (AI), a sub-branch of machine learning

I could have a failed mental model, but I'm under the impression that the relationship is the other way around. AI is a broad field encompassing various strategies to build intelligent machines. ML is one particular strategy where large volumes (think Big Data) of training data is used to teach by example. (Which makes Deep Learning a subset of ML, where "deep" neural networks are at play.)

Any news on the fast-neural-style release date?

What are the advantages you see over using the `go pprof` toolchain?

https://blog.golang.org/profiling-go-programs https://golang.org/pkg/net/http/pprof/ https://golang.org/pkg/runtime/pprof/