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

CuPy – NumPy-compatible matrix library accelerated by CUDA

cupy.chainer.org

Post reply on HN
Print view View original

1–10 of 36 posts

Re: CuPy – NumPy-compatible matrix library accelerated by CUDA

#3

Post by marmaduke »

fourier_mode wrote: post #2

Not sure what is the motivation behind this library. There are already several array GPU accelerated array libraries -- PyTorch, TensorFlow, ArrayFire, it even looks like pycuda has a small array class.

Did you see the « NumPy compatible » part of the title?

Re: CuPy – NumPy-compatible matrix library accelerated by CUDA

#5

Post by Smerity »

fourier_mode wrote: post #2

Not sure what is the motivation behind this library. There are already several array GPU accelerated array libraries -- PyTorch, TensorFlow, ArrayFire, it even looks like pycuda has a small array class.

Chainer and potentially CuPy (which was extracted from Chainer to be independent) were around before PyTorch as it served as inspiration for PyTorch. I feel like that's a good motivation for diversity in packages and ecosystems regardless of your feelings otherwise.

Along with a colleague I used CuPy in first Chainer and then PyTorch for implementing the Quasi-Recurrent Neural Network (QRNN) which at the time was far faster than even NVIDIA's optimized cuDNN LSTM whilst getting the same (or better) performance for many tasks.

CuPy at the time was both the easiest and most Pythonic of potential solutions for that problem - even if it did involve writing CUDA in Python strings =]

n.b. Our use case was literally pushing state of the art in research - CuPy is even more Pythonic if you're hitting more standard use cases.

[1]: https://github.com/salesforce/pytorch-qrnn

Re: CuPy – NumPy-compatible matrix library accelerated by CUDA

#7

Post by fourier_mode »

marmaduke wrote: post #3
fourier_mode wrote: post #2

Not sure what is the motivation behind this library. There are already several array GPU accelerated array libraries -- PyTorch, TensorFlow, ArrayFire, it even looks like pycuda has a small array class.

Did you see the « NumPy compatible » part of the title?

It is "highly compatible", similar statement can be made about other libs say torch tensors.

Re: CuPy – NumPy-compatible matrix library accelerated by CUDA

#8

Post by marmaduke »

fourier_mode wrote: post #7
marmaduke wrote: post #3

Earlier quoted context omitted.

Did you see the « NumPy compatible » part of the title?

It is "highly compatible", similar statement can be made about other libs say torch tensors.

Having tried to debug some issues between autograd, PyTorch & TensorFlow, I find torch & tf tensors have different enough syntax and naming that one needs to google a bit.
Post reply on HN