4 ms·

From the banner for "Data Analysis with Rust Notebooks" this looks someone realised that both data stuff and Rust are popular on their own and decided to put th

From the banner for "Data Analysis with Rust Notebooks" this looks someone realised that both data stuff and Rust are popular on their own and decided to put them together to sell an e-book. It's cool that this is at all possible but ultimately it forces Rust into a use case that it isn't well suited for. I can't see a good reason to pick it over Python/R/Julia.

whimsicalism 6y ago
Speed, static types. Definitely not a general alternative.
batterylow 6y ago
In the section before there is agreement with your comment: > Can we write and execute all our code in a Jupyter Notebook? Yes! Should we? Probably not. However, I enjoy the workflow, and making this an enjoyable process is important to me.
kyllo 6y ago
Yeah I think Rust's place in this ecosystem is a C/C++ replacement. All the Python and R data science packages call into native code to do the heavy lifting--the linear algebra, gradient descent etc. The next Tensorflow or PyTorch type framework could be implemented in Rust instead of C++. But it shouldn't matter to the end users--they would still use the Python or R bindings so they can have an interactive REPL environment to do their analysis in. Julia is interesting because it's performant enough to implement these things directly--Julia's machine learning frameworks are written in pure Julia (with the exception of the calls to CUDA libraries for GPU) and can achieve near-native performance, so solving the "two language problem" is part of its value proposition. I do wonder how valuable it is for the whole stack to be written in one language, and whether that will blur the lines between the software engineers who implement data science packages and the data scientists who use them.
- amkkma 6y ago
  Julia can do GPU kernels (Intel, AMD, Nvidia) also, they just aren't all implemented in pure Julia (or tuned) yet, hence the CUDA lib calls. https://juliagpu.org/ https://juliagpu.org/ https://juliagpu.org/2020-09-28-gemmkernels/ https://juliagpu.org/2020-09-28-gemmkernels/

whimsicalism 6y ago

Speed, static types. Definitely not a general alternative.

batterylow 6y ago

In the section before there is agreement with your comment: > Can we write and execute all our code in a Jupyter Notebook? Yes! Should we? Probably not. However, I enjoy the workflow, and making this an enjoyable process is important to me.

kyllo 6y ago

Yeah I think Rust's place in this ecosystem is a C/C++ replacement. All the Python and R data science packages call into native code to do the heavy lifting--the linear algebra, gradient descent etc. The next Tensorflow or PyTorch type framework could be implemented in Rust instead of C++. But it shouldn't matter to the end users--they would still use the Python or R bindings so they can have an interactive REPL environment to do their analysis in. Julia is interesting because it's performant enough to implement these things directly--Julia's machine learning frameworks are written in pure Julia (with the exception of the calls to CUDA libraries for GPU) and can achieve near-native performance, so solving the "two language problem" is part of its value proposition. I do wonder how valuable it is for the whole stack to be written in one language, and whether that will blur the lines between the software engineers who implement data science packages and the data scientists who use them.

amkkma 6y ago
Julia can do GPU kernels (Intel, AMD, Nvidia) also, they just aren't all implemented in pure Julia (or tuned) yet, hence the CUDA lib calls. https://juliagpu.org/ https://juliagpu.org/ https://juliagpu.org/2020-09-28-gemmkernels/ https://juliagpu.org/2020-09-28-gemmkernels/