Taylor Miller

Code coverage service: permissions and github integration

1

coveralls - [ ] configure account permissions - [ ] setup integration

## Implementation Ideas - [ ] calling `.ensemble()` could build an array of TSMs and then send them to `.roc_plot()` - [ ] the ensemble could return some kind of...

## Benefits - multiclass precision, recall, f1-score http://scikit-learn.org/stable/modules/generated/sklearn.metrics.classification_report.html#sklearn.metrics.classification_report

# Skatr - [ ] Try skatr on a real dataset - [ ] Write a blog post about it? https://github.com/datascienceinc/Skater

relates to https://github.com/HealthCatalyst/healthcareai-r/issues/642 https://github.com/mkleehammer/pyodbc/wiki/Connecting-to-SQL-Server-from-Mac-OSX

Set all n_jobs to -1 to use ALL THE CORES

1

> The power of scikit-learn’s multicore capabilities is shown online 8: In order to leverage all of the cores available on any system, we just pass the value -1 to...

## Background - This was not implemented for MVP reasons and for logging reasons (advanced users may not want their consoles full of stuff). ## Implementation ideas - Consider moving...

With the advent of `guess_positive_label()`, and the ability for a user to specify `binary_positive_label` when instantiating `TrainedSupervisedModel`, it's possible that they selected the incorrect class. This manifests as an inverted...

# Benefits - This could simplify and/or remove lots of code in feature importance, ROC, PR and confusion matrix plots - This would mean less code for us to maintain....

Initial Neural Net - DO NOT MERGE

3

This requires substantial review and discussion before merging.