Question: Future of RubixML/DL

[Anecho]

Hello,
How is the research on RubixML/DL#1 going?
Whole repository seems to be archived few years ago.. Will there ever be a GPU support on PHP with Tensor extension?
Or is there already an alternative to it?

Guys ar Tensorflow, according to https://discuss.tensorflow.org/t/tensorflow-for-php/1373 are not planning to implement it as independant wrapper. Instead they are advising to run TensorFlow Serving server for computing and interact with it through REST API

Would be glad to hear your thoughts on what is current globally best approach to utilize GPU while using PHP. Can't find much info on internet aswell, most posts are 4years old..

Multi-threading using RubixML/Tensor is a great oportunity, but CPU performance still can't compare to GPU, especially with specialised matrix multiplication cores from nvidia RTX.

[andrewdalpino]

Hey @Anecho thanks for the question. Yeah I had to put Rubix DL down to rest because I was the only developer and I had to choose between developing open-source and being homeless or working full-time and not being homeless. It's unfortunate because I feel PHP could really benefit from a library based on TensorFlow but alas reality is what it is.

Maybe someone else can revive the effort. There's a library called Rindow that I haven't had the chance to use yet, but from what I can see in the docs, it supports GPUs as well as some more advanced neural network layers.

https://github.com/rindow

[yarcowang]

@andrewdalpino Hi, I just found RubixML, great work!
But I installed it failed when it claimed "[-lgfortran](ld: library 'gfortran' not found)" after I setup openblas, lapacke.....
May I ask, can this (RubixML) be divided into two packages: classic ML, and DL?
Oh, I ask this question here because, as the title of this thread suggests, I recently feel PHP doesn't need DL but maybe needs DA.

(Maybe it could reduce the complexity and no need to reply on so many libs, am I right?)