Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Performance of ChainerX Linear Algebra routines

See original GitHub issue

Some chainerx.linalg routines (tracked at #6764) run more slowly compared to numpy.linalg and cupy.linalg. Performance results & code to obtain them is available in this gist.

The reason for slowness of qr, solve, inv is allocating temporary arrays in the kernel and copying data from them to routine level arrays. Also repeated transpose-copy operations (because of ChainerX’s row-major vs column-major ordering of LAPACK/cuSOLVER) have an effect. Routines that don’t require such manipulations run almost one to one with NumPy/CuPy.

Issue Analytics

State:
Created 4 years ago
Comments:5 (2 by maintainers)

Top GitHub Comments

1reaction

hvycommented, Aug 13, 2019

This is a linalg kernel design matter by the way. For the record, @IvanYashchuk and I have discussed this matter outside GitHub and decided to wait with addressing this issue until all ongoing linalg PRs have been merged. The reasons is that some routines are already merged, and they basically all share the same routine-kernel delegation

0reactions

stale[bot]commented, Apr 14, 2020

This issue is closed as announced. Feel free to re-open it if needed.

Top Results From Across the Web

Add Linear Algebra routines in ChainerX · Issue #6764 - GitHub

I'd like to work on adding linear algebra routines to ChainerX during GSoC. Most of the numpy.linalg functions listed in ChainerX routines ......

Ivan Yashchuk - Ivan Yashchuk

This summer I am going to work on extending ChainerX with some Linear Algebra routines. The goal is to implement common routines that...

Released Chainer v7.0.0b4 - Google Groups

Many updates to ChainerX including new routines and support for loss scaling. New Features. Support all float dtypes in F.n_step_rnn and F.

chainerx.huber_loss — Chainer 7.8.1 documentation

Element-wise Huber loss. The Huber loss is similar to the squared error but is less sensitive to outliers in the data. It is...

chainerx.hstack — Chainer 7.8.1 documentation

chainerx.hstack(arrays)¶. Stack arrays in sequence horizontally (column wise). Parameters. arrays (sequence of ndarray s) – Arrays to be stacked. Returns.