-
Notifications
You must be signed in to change notification settings - Fork 62
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
[ML] Guard against the case there are insufficient populated values to estimate variance #13
Conversation
…imate variance in tests for periodic components
LOG_TRACE(" autocorrelation = " << R); | ||
LOG_TRACE(" autocorrelationThreshold = " << Rt); | ||
if (R > Rt) | ||
if (df1 > 0.0) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Aside from this check, the following differences are all due to whitespace.
LOG_TRACE(" significance = " << CStatisticalTests::leftTailFTest(v1 / v0, df1, df0)); | ||
|
||
if (v1 <= vt && CStatisticalTests::leftTailFTest(v1 / v0, df1, df0) <= MAXIMUM_SIGNIFICANCE) | ||
if (df1 > 0.0) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Aside from this check, the following differences are all due to whitespace.
@hendrikmuhs can you have a quick look at this. This should ideally get merged before code freeze for 6.3. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
…imate variance in tests for periodic components (#13) This was triggering a boost exception to be thrown and spamming the logs. In all such cases, we have too little data to correctly determine if such a periodic component is present and so can correctly exit early.
…imate variance in tests for periodic components (#13) This was triggering a boost exception to be thrown and spamming the logs. In all such cases, we have too little data to correctly determine if such a periodic component is present and so can correctly exit early.
…imate variance in tests for periodic components (#13) This was triggering a boost exception to be thrown and spamming the logs. In all such cases, we have too little data to correctly determine if such a periodic component is present and so can correctly exit early.
... in tests for periodic components.
This was triggering a boost exception to be thrown and spamming the logs. In all such cases, we have too little data to correctly determine if such a periodic component is present and so can correctly exit early.
This shouldn't have any impact on results, since the test result should be the same, but should suppress log errors and is more efficient (although this code won't be a bottleneck).
Marking as a non-issue since this code has not been released.