Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why Swift for TensorFlow?

github.com

Post reply on HN
Print view View original

31–40 of 151 posts

Re: Why Swift for TensorFlow?

#31

Post by igotsideas »

yoz-y wrote: post #24
igotsideas wrote: post #8

Well written explanation! I really enjoy Swift but it's not as accessible as some of the other languages mentioned. I have a 2011 Macbook Pro and wanted to use the latest and greatest new Swift features. Unfortunately, my machine is too old to upgrade to Mojave which means I can't download the latest version of xcode, which means no new version of Swift. I'm not mad at Apple in the least bit. I just wish I could use…

If you don't mind stranding from the 100% stable roads, you can install Mojave on your macbook using this patcher: http://dosdude1.com/mojave/ Personally I am running Mojave on a late 2009 macbook pro and it still works amazingly well. Transition from Mojave and especially the new XCode are also way faster than previous iterations. There are caveats though, as the processor in my computer is too old, I had to hack ho…

I'm gonna try this, thank you!

Re: Why Swift for TensorFlow?

#32

Post by skwb »

dynamicwebpaige wrote: post #30
skwb wrote: post #26

Forgive me for my ignorance, but does swift have any good plotting and interactive "notebook" ability? Specifically the ability to plot images such as matplotlib. I ask this because the number 1 reason my deep learning research group chose python was because of the extensive and interactive scientific plotting ability that's built into python jupyter notebooks. While our volume of analysis isn't on the scale of say a…

Yes! Swift is supported in Google Colab, and as a Jupyter kernel: https://github.com/google/swift-jupyter .

What is the plotting experience like though? As I previously mentioned, plotting is one of the main reasons our group uses python.

Another reason now that I think about it, is the number of scientific libraries that I can just "pip install" without much thought (such as scipy/opencv).

Re: Why Swift for TensorFlow?

#34

Post by marmaduke »

3jckd wrote: post #29

It's interesting how it's going to play out. On one hand side, Swift is a pleasant language to work with (despite its infancy). But on the other, having a Tensorflow API doesn't suddenly give it a bunch of libraries for statistics, comp. vision, modeling, visualisation, etc. that Python/R/Julia cough MATLAB cough have. Nowadays, it's difficult enough to convince people to drop e.g. MATLAB for R or Python for Julia (l…

Julia doesn't tick the "compile to .o/.h" box. As far as I can tell, the use case for AOT Julia is avoiding package compilation overhead, not delivery of standalone code objects.

_edit_ seems JuliaC does support this sort of thing:

https://juliacomputing.com/blog/2016/02/09/static-julia.html

Re: Why Swift for TensorFlow?

#35

Post by aschampion »

3jckd wrote: post #29

It's interesting how it's going to play out. On one hand side, Swift is a pleasant language to work with (despite its infancy). But on the other, having a Tensorflow API doesn't suddenly give it a bunch of libraries for statistics, comp. vision, modeling, visualisation, etc. that Python/R/Julia cough MATLAB cough have. Nowadays, it's difficult enough to convince people to drop e.g. MATLAB for R or Python for Julia (l…

Yeah, I don't buy the justification versus Julia because of community size either, given most of Swift's community has little to do with data science. The document even says as much, contradicting that rationale, later on.

As someone who uses TF heavily, I would be much more excited about this project if they'd chosen Julia. Swift's tooling isn't great, and I already have a foot in one language with an immature data science ecosystem (Rust).

Re: Why Swift for TensorFlow?

#36

Post by marcrasi »

skwb wrote: post #32
dynamicwebpaige wrote: post #30

Earlier quoted context omitted.

Yes! Swift is supported in Google Colab, and as a Jupyter kernel: https://github.com/google/swift-jupyter .

What is the plotting experience like though? As I previously mentioned, plotting is one of the main reasons our group uses python. Another reason now that I think about it, is the number of scientific libraries that I can just "pip install" without much thought (such as scipy/opencv).

You can call out to matplotlib (or any other python libraries installed on your system), using the python interop feature (https://github.com/tensorflow/swift/blob/master/docs/PythonI...)!

https://github.com/google/swift-jupyter#rich-output has an example with screenshots.

Re: Why Swift for TensorFlow?

#37

Post by kodablah »

Strange to see a requirement for choosing one language over another is supposed ease of adoption and then they choose the one not easily adopted across every platform. That easy to write syntax takes precedence over general easy to write/run is unfortunate.

Re: Why Swift for TensorFlow?

#38

Post by phillipcarter »

coder543 wrote: post #21
criddell wrote: post #20

Earlier quoted context omitted.

From the article: We believe that Rust supports all the ingredients necessary to implement the techniques in this paper: it has a strong static side, and its traits system supports zero-cost abstractions which can be provably eliminated by the compiler. It has a great pointer aliasing model, a suitable mid-level IR, a vibrant and engaging community, and a great open language evolution process. A concern with using Ru…

As I pointed out in two lengthy comments on day one[1][2], that reasoning is nonsense. If Chris wants to use the language he created in this new endeavor for machine learning simply because he made it , that's totally fine and completely his prerogative, but he should just say so, rather than trying (and failing) to convince people that other languages aren't better suited for this task. From my point of view, a weak…

I personally find Rust to have quite a learning curve (which I guess is also an opinion shared by others). The language is great though.

I do agree with your criticism of the document here, though. It feels very much like Swift happens to check many boxes, but the lack of Windows support is baffling. It's simply table stakes to be able to run, fully supported, on Windows, macOS, and major Linux distributions. That should be the very first thing anyone considers.

But beyond that, I think even with Rust's macro system it could be difficult to make it work for Tensorflow in a way that feels appropriate for Rust programmers _and_ for TensorFlow. This was explored in F# for Tensorflow research[0] and a completely different approach[1] was taken because making a type system suitable for tensorflow got too unweildy.

[0]: https://github.com/fsprojects/TensorFlow.FSharp

[1]: https://github.com/fsprojects/TensorFlow.FSharp#live-checkin...

Re: Why Swift for TensorFlow?

#39

Post by zozbot123 »

elpakal wrote: post #5

I also wonder how much of this coincidentally lines up with Chris Lattner landing at Google. As Chris will admit, and as was left out of this analysis, Swift has also been given the humble goal of achieving world domination. All joking aside I'm very thrilled about this and have enjoyed tremendously watching the Swift language mature since its launch due in large part to the open source community and the Swift team's…

> ...Swift has also been given the humble goal of achieving world domination

Unfortunately it's too late for Swift - the Rust Evangelism Strike Force (aided by the Rewrite-It-In-Rust Task Force) is well on the way towards achieving this goal for Rust, all by themselves! There is no place remaining for Swift, D or any other language in the same niche.

Re: Why Swift for TensorFlow?

#40

Post by rudedogg »

m0zg wrote: post #7

Is there a "standard" way of running Swift on Ubuntu LTS nowadays? A while back I looked into it, and ran into some hokey and unsatisfying solutions. I used Swift on iOS, and I like it a lot, but if they care about adoption, someone needs to reduce friction of getting up and running to approximately zero. A snap package a-la Go or per-user script based installation a-la Rust would be quite OK, as long as it's just on…

It looks pretty straightforward, see the Linux section of https://swift.org/download/#using-downloads
Post reply on HN