Wikipedia:Articles for deletion/PlaidML

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was no consensus (WP:NPASR). King of ♥ ♦ ♣ ♠ 01:26, 25 March 2019 (UTC)[reply]

PlaidML[edit]

[Hide this box] New to Articles for deletion (AfD)? Read these primers!

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

I've actually come across PlaidML as part of my day job, but it doesn't appear to me that there's enough coverage in reliable sources for it to meet notability guidelines. In an internet search for the term, I was able to find a few Medium blog posts that examined PlaidML in depth, but no RS. A Google Scholar search returned a half dozen articles that name-dropped PlaidML as an example of a Keras-compatible backend, but don't appear to give it further coverage (and none of them had significant citation counts). Does not meet WP:GNG, WP:NSOFTWARE. _signed,Rosguill ^talk 02:28, 4 March 2019 (UTC)[reply]

- I respectfully disagree with you on this. PlaidML is an important piece of software as it's currently the only way to circumvent the CUDA monopoly and perform platform-independent machine learning tasks without having to rely on a single vendor and its proprietary framework. I'm well aware of the fact that both PlaidML AND this article still need a lot of work and have lots of room for improvements but deleting it would definitely do more harm than good. Thus, I'm strongly in favour of keeping it. Gromobir (talk) 09:50, 7 March 2019 (UTC)[reply]

Do you have sources that would back up those statements? You may be right about the role of PlaidML among other ML tools, but ultimately we need reliable sources saying this that we can cite. _signed,Rosguill ^talk 17:54, 7 March 2019 (UTC)[reply]

Of course, there references backing up my statements. For example:

"The PlaidML really surprised me with its ease of installation, performance and substantial documentation.", "I must say that this project not only makes the domain of deep learning accessible to a wider variety of people but also helps reduce the demand, and hopefully the price, of GPUs." (https://medium.com/@Vatsal410/keras-without-nvidia-gpus-with-plaidml-and-amd-gpu-4ba6f60025ce).
"If, however, you have an AMD GPU card, as I do in my University-provided 2017 Macbook Pro, then none of the above support your hardware, you have very few options. In fact, until this blog post, I thought you had none. Enter PlaidML! PlaidML is another machine learning engine – essentially a software library of neural network and other machine learning functions. Conveniently, PlaidML can be used as a back-end for Keras also. And, unlike basically every other such engine, PlaidML is designed for OpenCL, the poorer, open-source cousin of NVIDIA’S CUDA GPU programming language. Plus, it works on Macs. (PlaidML is Python based)." (https://informatics.sydney.edu.au/blogs/amdgpu/)
"PlaidML is a multi-language acceleration framework that:

   Enables practitioners to deploy high-performance neural nets on any device,
   Allows hardware developers to quickly integrate with high-level frameworks,
   Allows framework developers to easily add support for many kinds of hardware,
   Works on all major platforms — Linux, macOS, Windows.,
   Accelerates deep learning on AMD, Intel, NVIDIA, ARM, and embedded GPUs."

(https://medium.com/@akashdeepjassal/plaidml-a-alternative-open-source-deep-learning-library-for-all-gpus-accfe6b879b)

Do you consider these sources to be reliable? — Preceding unsigned comment added by Gromobir (talk • contribs) 15:04, 8 March 2019 (UTC)[reply]

Medium posts are generally not reliable unless the author is a readily identifiable expert in the given subject (see WP:BLOGS). The Sydney University post, despite being a blog, is likely reliable due to its affiliation with a reputable research institution. However, I'm not sure that that article has enough information about the subject to single handedly establish notability: all the article tells us is that 1) PlaidML can be run on a 2017 Macbook Pro and 2) how to install it. Admittedly #1 does make it somewhat unique among other ML toolkits, but ideally I'd like to see more coverage about its actual performance. _signed,Rosguill ^talk 18:23, 8 March 2019 (UTC)[reply]

When it comes to the performance, some Phoronix benchmarks are available: (https://www.phoronix.com/scan.php?page=article&item=plaidml-nvidia-amd&num=1). Gromobir (talk) 11:09, 11 March 2019 (UTC)[reply]

Note: This discussion has been included in the list of Software-related deletion discussions. Icewhiz (talk) 07:19, 4 March 2019 (UTC)[reply]

Keep It may be controversial but it is a fairly obscure and at this time unique product that it is worth keeping the article. Its bleeding edge software design and WP has few of these types of articles. scope_creep^Talk 12:12, 9 March 2019 (UTC)[reply]

Relisted to generate a more thorough discussion and clearer consensus.
Please add new comments below this notice. Thanks, Xymmax ^{So let it be written} _{So let it be done} 18:06, 11 March 2019 (UTC)[reply]

Relisted to generate a more thorough discussion and clearer consensus.
Relisting comment: Given both the split views and particularly on the status of the given sources
Please add new comments below this notice. Thanks, Nosebagbear (talk) 22:49, 18 March 2019 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.