Daft equivalent of polars-lts-cpu

[jpedrick-numeus]

Is your feature request related to a problem?

I often develop for an x86 environment on an M1 MacBook using an x86_64 docker container. Related to #1021 as I consider adding Daft to my technology stack it causes some issues when testing/building docker images that incorporate Daft.

Describe the solution you'd like

Polars solves this problem by providing a package polars-lts-cpu which runs ok on an M1 chipset inside an x86_64 docker container. I don't have any strong opinions on the particular solution. As an end user here I would only like to be able to run Daft on an x86_64 VM running on an M1 MacBook in a docker container.

Describe alternatives you've considered

Developing on a remote x86_64 based EC2 instance

Additional Context

No response

Would you like to implement a fix?

No

[andrewgazelka]

Hello @jpedrick-numeus can you specify which issues specifically you are experiencing? Is there some instruction set (say SIMD) that the x86 emulator you are using is not supoprting? Thank you :)

[jpedrick-numeus]

Hi @andrewgazelka

Hello @jpedrick-numeus can you specify which issues specifically you are experiencing? Is there some instruction set (say SIMD) that the x86 emulator you are using is not supoprting? Thank you :)

I'm not an expert regarding instruction sets. The module crashes on import.

$ python 
Python 3.10.15 | packaged by conda-forge | (main, Oct 16 2024, 01:24:24) [GCC 13.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import daft
Illegal instruction

I'm running Docker Machine on an M1 MacBook w/ an x86_64 docker image (mambaorg/micromamba) w/ the Rosetta emulator.