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

PyTorch 2.0

pytorch.org

Post reply on HN
Print view View original

91–100 of 111 posts

Re: PyTorch 2.0

#91

Post by ldjkfkdsjnv »

vikinghckr wrote: post #70
hintymad wrote: post #25

A big lesson I learned from PyTorch vs other frameworks is that productivity trumps incremental performance improvement. Both Caffe and MXNet marketed themselves for being fast, yet apparently being faster here and here by some percentage simply didn't matter that much. On the other hand, once we make a system work and make it popular, the community will close the performance gap sooner than competitors expect. Anoth…

What I really find interesting here is that PyTorch, a library maintained by Facebook, is winning the marketshare and mindshare due to clean API, whereas Tensorflow, maintained by Google, is losing due to inferior API. In general, Google as a company emphasizes code quality and best practices far more than Facebook. But the story was reversed here.

The same thing is true of react vs angular. Angular is kind of garbage, whereas react is elegant

Re: PyTorch 2.0

#92

Post by chem83 »

aschleck wrote: post #90
vikinghckr wrote: post #70

Earlier quoted context omitted.

What I really find interesting here is that PyTorch, a library maintained by Facebook, is winning the marketshare and mindshare due to clean API, whereas Tensorflow, maintained by Google, is losing due to inferior API. In general, Google as a company emphasizes code quality and best practices far more than Facebook. But the story was reversed here.

It's worth noting that Google is also working on JAX these days and that's picking up steam, and one reason is because its API is super clean.

And ML compilers in the form of IREE/MLIR, which if successful in delivering hw and framework optionality, should make ML frameworks a far less consequential decision.

Re: PyTorch 2.0

#93

Post by orbital-decay »

qayxc wrote: post #82
anigbrowl wrote: post #79

I get that Nvidia is the favorite GPU (because CUDA) and that library maintainers want to chase the latest and best to do the most. But I don't get why support for older hardware (including CUDA stuff) is just deprecated or abandoned, nor why support for other GPU architectures is lacking across many popular ML libraries. A lot of this is down to driver availability and software stacks...but is all of it? Game engine…

NVIDIA invested billions in their software infrastructure over the past 15 years (CUDA was first released in 2007) and basically brute forced their way into academia back in 2008, sponsoring labs around the world to get GPGPU going. Both mindshare and market share (outside of super computers) are just overwhelming at this point. Their market share in the consumer market is ~86% as of this quarter. The data centre mar…

> NVIDIA invested billions in their software infrastructure over the past 15 years (CUDA was first released in 2007) and basically brute forced their way into academia back in 2008, sponsoring labs around the world to get GPGPU going.

Worth mentioning that CUDA didn't appear in vacuum either - GPGPU was a thing since the first shaders in consumer GPUs, which were also introduced by NVidia in NV20. CUDA was a result of years of ad-hoc attempts at GPU programming. So they really just built the entire field from nothing.

Re: PyTorch 2.0

#94

Post by perturbation »

antixk wrote: post #86
la_fayette wrote: post #53

Earlier quoted context omitted.

:) it still is impossible to bring a cnn rnn network in pytorch to mobile, which works fine with tflite...

Not sure if you have heard of Pytorch Mobile but it is very possible[0] [0] https://pytorch.org/tutorials/beginner/deeplabv3_on_ios.html

The big thing that PyTorch Mobile is lacking compared to TF Lite is on-device accelerator support (GPU/DSP/etc.) (there's experimental support for NNAPI https://pytorch.org/tutorials/prototype/nnapi_mobilenetv2.ht..., but this is a hack).

Re: PyTorch 2.0

#95

Post by pjmlp »

ShamelessC wrote: post #84
vikinghckr wrote: post #70

Earlier quoted context omitted.

What I really find interesting here is that PyTorch, a library maintained by Facebook, is winning the marketshare and mindshare due to clean API, whereas Tensorflow, maintained by Google, is losing due to inferior API. In general, Google as a company emphasizes code quality and best practices far more than Facebook. But the story was reversed here.

Google historically has a culture of writing libraries with other Google employees as the target audience.

The way Android NDK and AGK experience works, I always have the feeling it is maintained as a 20% project and none of the developers has ever used anything from the competition.

Re: PyTorch 2.0

#96

Post by pjmlp »

synergy20 wrote: post #30
quietbritishjim wrote: post #5

> Today, we announce torch.compile, a feature that pushes PyTorch performance to new heights and starts the move for parts of PyTorch from C++ back into Python. I'll admit I don't know enough about PyTorch to know what torch.compile is exactly. But does this means some features of PyTorch will no longer be available in the core C++ library? One of the nice things about PyTorch had been that you could do your training…

Facebook did similar thing to its original code PHP, it uses HHVM to 'compile' PHP(now called Hacklang) to gain performance, it seems doing similar thing with python here.

If I remember correctly, the JIT version of the PHP compiler proved itself against the AOT compilation to native code via translation to C++, so they went with the more productive variant.

Re: PyTorch 2.0

#97

Post by mdda »

anigbrowl wrote: post #85
qayxc wrote: post #82

Earlier quoted context omitted.

NVIDIA invested billions in their software infrastructure over the past 15 years (CUDA was first released in 2007) and basically brute forced their way into academia back in 2008, sponsoring labs around the world to get GPGPU going. Both mindshare and market share (outside of super computers) are just overwhelming at this point. Their market share in the consumer market is ~86% as of this quarter. The data centre mar…

Thanks much for this in-depth explanation. I've been struggling with this for a while as I am low on the learning curve with a lot of ML stuff (partly due to headaches with finding a GPU that was affordable and properly supported in order to develop elementary competence). I think I'm just going to get an eGPU with a recent & decent CUDA card rather than waiting for a utopia of interoperability and backwards compatib…

Google Colab gives you $free GPU (usually a 16Gb T4) preloaded with frameworks, ready to run. Later, you might be tempted by the Pro(+) version, but there's plenty of scope to move up the learning curve before spending any money.

Re: PyTorch 2.0

#98

Post by anigbrowl »

mdda wrote: post #97
anigbrowl wrote: post #85

Earlier quoted context omitted.

Thanks much for this in-depth explanation. I've been struggling with this for a while as I am low on the learning curve with a lot of ML stuff (partly due to headaches with finding a GPU that was affordable and properly supported in order to develop elementary competence). I think I'm just going to get an eGPU with a recent & decent CUDA card rather than waiting for a utopia of interoperability and backwards compatib…

Google Colab gives you $free GPU (usually a 16Gb T4) preloaded with frameworks, ready to run. Later, you might be tempted by the Pro(+) version, but there's plenty of scope to move up the learning curve before spending any money.

I should check that out. Jetbrains just integrated remote management for code and notebooks into their IDEs and this seems like the perfect way to test. Thanks for the tip!

Re: PyTorch 2.0

#99

Post by rawrawrawrr »

chem83 wrote: post #92
aschleck wrote: post #90

Earlier quoted context omitted.

It's worth noting that Google is also working on JAX these days and that's picking up steam, and one reason is because its API is super clean.

And ML compilers in the form of IREE/MLIR, which if successful in delivering hw and framework optionality, should make ML frameworks a far less consequential decision.

Not really, because people will still use the dominant framework because that's where all the research, examples, libraries are.

Re: PyTorch 2.0

#100

Post by rawrawrawrr »

vikinghckr wrote: post #70
hintymad wrote: post #25

A big lesson I learned from PyTorch vs other frameworks is that productivity trumps incremental performance improvement. Both Caffe and MXNet marketed themselves for being fast, yet apparently being faster here and here by some percentage simply didn't matter that much. On the other hand, once we make a system work and make it popular, the community will close the performance gap sooner than competitors expect. Anoth…

What I really find interesting here is that PyTorch, a library maintained by Facebook, is winning the marketshare and mindshare due to clean API, whereas Tensorflow, maintained by Google, is losing due to inferior API. In general, Google as a company emphasizes code quality and best practices far more than Facebook. But the story was reversed here.

Google has a culture of encouraging their developers to build complex things, which is important for promotion.
Post reply on HN