search

rapidsai / cuml

cuML - RAPIDS Machine Learning Library
https://docs.rapids.ai/api/cuml/stable/
Apache License 2.0
4.26k stars 535 forks source link

[QST] Can cuml and cudf installed on nvidia jetson tx1/tx2/nano? #665

Open lk1983823 opened 5 years ago

lk1983823 commented 5 years ago

What is your question? I know that anaconda can be installed using Archiconda. However, I dont know wheter cuml and cudf can installed on nvidia jetson tx1/tx2/nano? Or does these two packages have aarch64 version?

taureandyernv commented 5 years ago

Sadly no (technically not yet, and it probably wouldn't be officially supported). RAPIDS, and its libraries like cuML, require Pascal architecture and higher. The Jetson boards use Maxwell Architecture. RAPIDS does work on the Xavier boards.

nagarjunredla commented 5 years ago

The Jetson TX2 uses Pascal architecture. I've got cudf to install on the TX2 with workarounds similar to the ones for Xavier and I'm currently trying to install cuml. It would be helpful if there was some more support. I still have a few dependency issues, will post once I have updates.

lk1983823 commented 5 years ago

The Jetson TX2 uses Pascal architecture. I've got cudf to install on the TX2 with workarounds similar to the ones for Xavier and I'm currently trying to install cuml. It would be helpful if there was some more support. I still have a few dependency issues, will post once I have updates.

Thanks for your reply. Hope you can keep updating.

tkpudgy commented 3 years ago

@nagarjunredla Hello, were you able to install CuML on a Jetson and use it? I have a Jetson Xavier NX. I would like to install CuML on it. Any help would be appreciated. Thank you.