binaryClassifierEmailReport: Email a Binary Classifier Report
In ivanliu1989/RQuant: A collection of generic functions

Description Usage Arguments Value See Also Examples

Given the result of binaryClassifierEvaluation email a standardised report to one or more recipients.

The emailed report contains a histogram of estimated probabilities, the ROC plot with the calculated area under the curve, two density plots, one violin, one standard, two loss curves as a function of deciles of the calculated probabilies, and a plot of cumulative It contains some stats such as Area under the ROC curve, Root Mean Square Error and the confusion matrix. It does not require the actual model. If the model is supplied, a summary of the model is also included in the output.

1
2
3

binaryClassifierEmailReport(evaluationResults,
  toAddress = "your.name@gmail.com", additionalMessage = "", plotSize = 10,
  emailFrontEnd = "")

`evaluationResults`	The result from running `binaryClassifierEvaluation`.
`toAddress`	Character vector, containing the email address(es) to send the report to, when completed.
`additionalMessage`	A string that will appear in the subject of the email
`plotSize`	Numeric scalar containing the side length of the resulting square png files, in pixels.
`emailFrontEnd`	A string of html that goes at the front of the email

The result of the call to sendmail

RQuantSendMail, binaryClassifierEvaluation

set.seed(123)
probs <- pnorm(rnorm(100,-1,1))
hist(probs)
actuals <- rbinom(100,1,probs)
plot(as.factor(actuals),probs)

binaryClassifierEmailReport(binaryClassifierEvaluation(probs, actuals),"ivan@xxx.com")