Skip to main content

A memray benchmark plugin for asv

Project description

asv_bench_memray

This a proof-of-concept externally defined memray benchmark plugin for asv.

Usage

In your asv.conf.json add the following to the requirements matrix:

"matrix": {
    "req": {
        "numpy": ["1.21"],
        "asv_plugin_memray": [""]
    },
},

This should be enough for an example like the following to run:

import numpy as np

class MyBenchmark:
    params = [10, int(2e4)]

    def ray_sum(self, n):
        self.data = np.random.rand(n)
        np.sum(self.data)

Which should result in something roughly like:

ASV_RUNNER_PATH="../asv_runner" asv run --bench sum
· Creating environments.......
· Discovering benchmarks
·· Uninstalling from virtualenv-py3.9-asv_bench_memray.._asv_bench_memray-numpy1.21
·· Building 30a650cb <main> for virtualenv-py3.9-asv_bench_memray.._asv_bench_memray-numpy1.21.
·· Installing 30a650cb <main> into virtualenv-py3.9-asv_bench_memray.._asv_bench_memray-numpy1.21
· Running 1 total benchmarks (1 commits * 1 environments * 1 benchmarks)
[ 0.00%] · For project commit 30a650cb <main>:
[ 0.00%] ·· Benchmarking virtualenv-py3.9-asv_bench_memray.._asv_bench_memray-numpy1.21
[100.00%] ··· benchmarks.MyBenchmark.ray_sum                                                ok
[100.00%] ··· ======== =======
               param1         
              -------- -------
                 10     1.72k 
               20000     162k 
              ======== =======

ASV_RUNNER_PATH="../asv_runner" asv run --bench sum  8.72s user 3.04s system 99% cpu 11.877 total

Development

During development, one needs to pass a relative path:

"matrix": {
    "req": {
        "asv_bench_memray": ["../asv_bench_memray"]
    },
},

Also it may be necessary to cleanup older environments and explicitly point to an instance of asv_runner:

rm -rf .asv && ASV_RUNNER_PATH="../asv_runner" asv run --bench sum

Personally I keep benchmarks for testing at an experiments repo, but any local test will do.

Conventions

Like all externally defined benchmark plugins for asv, this has a strict hierarchy.

  • The package name begins with asv_bench
  • Benchmarks are defined in a benchmarks folder under the package module
  • Each exported new benchmark type has the export_as_benchmark = [NAMEBenchmark] attribute

For more conventions follow the documention of asv_runner.

License

MIT.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

asv_bench_memray-0.1.2.tar.gz (10.4 kB view details)

Uploaded Source

Built Distribution

asv_bench_memray-0.1.2-py3-none-any.whl (4.2 kB view details)

Uploaded Python 3

File details

Details for the file asv_bench_memray-0.1.2.tar.gz.

File metadata

  • Download URL: asv_bench_memray-0.1.2.tar.gz
  • Upload date:
  • Size: 10.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: pdm/2.12.2 CPython/3.9.18

File hashes

Hashes for asv_bench_memray-0.1.2.tar.gz
Algorithm Hash digest
SHA256 f00c552ee20c72c4885d70d7066da043855c62218ffbf2ac0f3f3c9bd99e8d2c
MD5 06216e3e0f8929f8246eaed68f587bb9
BLAKE2b-256 9cda8b72a56c79131501f5466e33121d79fd9804059e7ad4164a4aa0aaac1e61

See more details on using hashes here.

Provenance

File details

Details for the file asv_bench_memray-0.1.2-py3-none-any.whl.

File metadata

File hashes

Hashes for asv_bench_memray-0.1.2-py3-none-any.whl
Algorithm Hash digest
SHA256 5bde1fd68f75317f9f9d8129bc9360734cced91cf1ea7334559b21d305157f2f
MD5 622df27be675fecbc91f31953d952bfd
BLAKE2b-256 df4f16ae7ad03d54e8a2c5a9cd20bd1a13471713b74e4c30c5375e5df184e033

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page