Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix unstable results in hist method #4767

Closed
wants to merge 1 commit into from
Closed

Fix unstable results in hist method #4767

wants to merge 1 commit into from

Conversation

SmirnovEgorRu
Copy link
Contributor

After my changes I observed some small differences in results from run to run.
For example, log-loss for HIGGS data set:
1 run: 0.807342768946606615
2 run: 0.807342768708895875

The reason was so easy - guided scheduling instead of static in OpenMP loop. It leads to non-deterministic operations with floating-point numbers.

I have fixed it and now log-loss:
1 run: 0.807342767697801778
2 run: 0.807342767697801778
3 run: 0.807342767697801778

Training time is the same as it were before:
Before time: 2951ms
After time: 2903ms

@SmirnovEgorRu
Copy link
Contributor Author

Ci contains following error:
tests/python/test_linear.py::TestLinear::test_shotgun terminate called after throwing an instance of 'std::bad_alloc'
[2019-08-13T11:59:10.252Z] what(): std::bad_alloc

I don't sure, that it is my problem, probably some infrastructure issue. Can we restart this?

@hcho3
Copy link
Collaborator

hcho3 commented Aug 13, 2019

See #4716, which already removed guided scheduling.

@hcho3 hcho3 closed this Aug 13, 2019
@lock lock bot locked as resolved and limited conversation to collaborators Nov 11, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants