undefined | Better HN

0 pointsvegabook9y ago0 comments

Thanks for the interesting and informative comment. Do I sense just a tiny bit of regret though? Yet another Python interface. YAPI. You heard it here first. And no, Py3 is not that nice. Too much cruft by far. And lua is miles faster than Python when you're outside the tensor domain, ie while you're sourcing and wrangling your data. Arguably luajit obviates the need for C , something you can't say about Python. Disclosure: I am a massive, but increasingly disenchanted, user of Python. I had actually started looking at Torch7, foregoing tensorflow, precisely because of Lua. But the walls are closing in....

0 comments

jjawssd9y ago

A very large portion of performance problems can be mitigated with the use of cython and the new asyncio stuff.

asyncio success story: https://magic.io/blog/asyncpg-1m-rows-from-postgres-to-pytho...

cython: http://scikit-learn.org/stable/developers/performance.html

vegabookOP9y ago

Luajit is at least 10x faster than python and easily obviates the need to mess around with cython. That's an easy win for Lua. Let's be honest: Torch has decided that if you cannot beat them, join them. It is about network effects. Not about Python better than Lua intrinsically.

baq9y ago

but why do you care if luajit is faster than python if everything that matters is computed on the GPU anyways?

jjawssd9y ago

Can't argue with that

inlineint9y ago

An alternative to Cython is Numba [1], which speeds up some cycles in pure Python by just adding a single decorator.

[1] http://numba.pydata.org/

cr0sh9y ago

> And lua is miles faster than Python when you're outside the tensor domain, ie while you're sourcing and wrangling your data.

Then use Lua for that, if you are more comfortable there and want/need the speed bump. There's nothing that says an entire project or whatnot has to be developed in a singular language.

Use each tool to its strengths, as your needs, requirements, and abilities dictate.

vegabookOP9y ago

There always has to be someone rolling out the horses-for-courses pitch. No. I wanted Lua to gain traction with other people. That's the point. I would have liked the Lua sci-ecosystem to be healthy as an alternative.

infinite8s9y ago

Is there an equivalent to numpy in the Lua space?

1 more reply

webmaven9y ago

> And lua is miles faster than Python when you're outside the tensor domain, ie while you're sourcing and wrangling your data.

Is that true even if the Python used is PyPy rather than CPython?

j / k navigate · click thread line to collapse

0 pointsvegabook9y ago0 comments

0 comments

jjawssd9y ago

A very large portion of performance problems can be mitigated with the use of cython and the new asyncio stuff.

asyncio success story: https://magic.io/blog/asyncpg-1m-rows-from-postgres-to-pytho...

cython: http://scikit-learn.org/stable/developers/performance.html

vegabookOP9y ago

baq9y ago

but why do you care if luajit is faster than python if everything that matters is computed on the GPU anyways?

jjawssd9y ago

Can't argue with that

inlineint9y ago

An alternative to Cython is Numba [1], which speeds up some cycles in pure Python by just adding a single decorator.

[1] http://numba.pydata.org/

cr0sh9y ago

> And lua is miles faster than Python when you're outside the tensor domain, ie while you're sourcing and wrangling your data.

Then use Lua for that, if you are more comfortable there and want/need the speed bump. There's nothing that says an entire project or whatnot has to be developed in a singular language.

Use each tool to its strengths, as your needs, requirements, and abilities dictate.

vegabookOP9y ago

infinite8s9y ago

Is there an equivalent to numpy in the Lua space?

1 more reply

webmaven9y ago

> And lua is miles faster than Python when you're outside the tensor domain, ie while you're sourcing and wrangling your data.

Is that true even if the Python used is PyPy rather than CPython?

j / k navigate · click thread line to collapse