Skip to content
This repository was archived by the owner on May 29, 2023. It is now read-only.

Releases: Paperspace/gradient-utils

v0.5.0

22 Oct 17:58
8fa97e4

Choose a tag to compare

What's Changed

Full Changelog: v0.3.2...v0.5.0

v0.3.2

07 Dec 19:26
5259ad9

Choose a tag to compare

  • Downgrades numpy and hyperopt to be compatible with Tensorflow and gradient-sdk

v0.3.1

19 Nov 22:18
c0d7cfa

Choose a tag to compare

  • Removes add_metrics function
  • Downgrades numpy for backwards compatability

v0.3.0

13 Nov 22:56
64b0b3b

Choose a tag to compare

Adds add_metrics function to easily push metrics in a Gradient environment

v0.2.0

14 Sep 15:17
3f77e32

Choose a tag to compare

Custom metrics support for notebooks and model deployments.

v0.1.5

08 Sep 21:17
b2d702c

Choose a tag to compare

Merge pull request #18 from BartoszCki/Fix-labels-of-metrics

Fix labels so it's visible to the API

Hostname parsing fix

18 Jun 14:31
d6d9fe9

Choose a tag to compare

Merge pull request #16 from Paperspace/hostname-parsing-fix

Hostname parsing fix

v0.1.2: Merge pull request #12 from Paperspace/fixing-numpy-version

22 May 14:39
b579e80

Choose a tag to compare

Release 0.1.0

13 May 12:07

Choose a tag to compare

MetricsLogger

Prometheus wrapper for logging custom metrics

Usage example:

from gradient_utils import MetricsLogger
m_logger = MetricsLogger()
m_logger.add_gauge("some_metric_1")
m_logger["some_metric_1"].set(3)
m_logger["some_metric_1"].inc()

m_logger.add_gauge("some_metric_2")
m_logger["some_metric_2"].set_to_current_time()

m_logger.push_metrics()