I wrote a thing about high-performance machine learning in @rustlang!
Honestly I found it surprising how few resources exist regarding SIMD, BLAS, and Lapack, considering both their history and how critical they are to MLhttps://www.erikpartridge.com/2019-03/rust-ml-simd-blas-lapack …
Curious about your previous experiences with ndarray. What have you tried to accomplish that turned out to be complicated/unintuitive? Do you have any specific routine where there was a considerable difference between the ndarray version and the low-level BLAS/LAPACK version?
-
-
About four months ago now, I took a shot at a simple neural net in each of the major lin alg libs to get to know them. I'll check if I still have the code, I'm not sure I do :(
-
ndarray is indubitably nicer than working with BLAS—I recall the biggest issue I had was building a generic and simple outward facing API for my neural net using ndarray. I use it for some of my non-public projects and I like it, but haven't tried another public API since.
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
Writing about stuff to learn how it works, mostly in Rust.
Lead Engineer at