Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
After extensive exploration, it was determined that the
overall
error rates were consistent across the two functions (perf.mint.splsda()
andtune.mint.splsda()
). Hence, only theBER
calculation needed to be rectified.The inconsistency was caused by the way the global BER was calculated in each function:
tune.mint.splsda()
: Calculated the confusion matrix of a model for each study and then found the mean of these study-specific BER valuesperf.mint.splsda()
: Calculated the confusion matrix of a model on all samples, irrespective of study. A single BER was yielded from this.The
perf.mint.splsda()
had its functionality changed such that it was brought in line with the way in whichtune.mint.splsda()
calculated global BER.The way in which
global$overall
is calculated inperf.mint.splsda()
was only adjusted for consistency withglobal$BER
- the outputted values in the previousoverall
method and newoverall
method are the same.A minor addition was made to BOTH functions (in the case of
tune.mint.splsda()
, this occured inLOGOCV()
). Rather than the global BER being just a basic average of each study-specific BER, both now find the average BER across studies, weighted by the study sample sizes