/// <summary>
/// Gets the <see cref="BiasedCommunityModelRunner" /> results.
/// </summary>
/// <param name="runner">
/// The runner.
/// </param>
/// <returns>
/// The dictionary of results.
/// </returns>
public static Dictionary <string, object> GetBiasedCommunityTrainingResults(ModelRunnerBase runner)
{
var result = GetModelTrainingResults(runner);
if (runner is BiasedCommunityModelRunner biasedCommunityRunner)
{
var communityCpts = new Dictionary <string, Dirichlet[]>();
for (var i = 0; i < biasedCommunityRunner.CommunityCpt.Count; i++)
{
communityCpts[i.ToString()] = biasedCommunityRunner.CommunityCpt[i];
}
var plottableCommunityCpts = PlotData.GetWorkerCpts(
communityCpts,
runner.DataMapping);
var communityCptsPerc = PlotData.GetWorkerCpts(
communityCpts,
runner.DataMapping,
true);
result["CommunityCpts"] = plottableCommunityCpts;
result["CommunityCptsPerc"] = communityCptsPerc;
result["WorkerCommunities"] = biasedCommunityRunner.WorkerCommunities;
result["WorkerCommunityCounts"] = biasedCommunityRunner.WorkerCommunityCounts;
}
return(result);
}
/// <summary>
/// Gets the <see cref="BiasedWorkerModelRunner" /> results.
/// </summary>
/// <param name="runner">
/// The runner.
/// </param>
/// <param name="maximumNumberWorkers">
/// The maximum number of workers for whom to return results..
/// </param>
/// <returns>
/// The dictionary of results.
/// </returns>
public static Dictionary <string, object> GetBiasedWorkerTrainingResults(ModelRunnerBase runner,
int maximumNumberWorkers = 20)
{
var result = GetModelTrainingResults(runner);
if (runner is BiasedWorkerModelRunner biasedWorkerRunner)
{
var prominentWorkers = GetProminentWorkers(runner.DataMapping.Data, maximumNumberWorkers);
var prominentWorkerCpts = biasedWorkerRunner.WorkerCpt
.Where(kvp => prominentWorkers.Contains(kvp.Key)).ToDictionary(kvp => kvp.Key, kvp => kvp.Value);
var workerCpts = PlotData.GetWorkerCpts(
prominentWorkerCpts,
runner.DataMapping);
var workerCptsPerc = PlotData.GetWorkerCpts(
prominentWorkerCpts,
runner.DataMapping,
true);
result["ProminentWorkerCpts"] = workerCpts;
result["ProminentWorkerCpts"] = workerCptsPerc;
}
return(result);
}
/// <summary>
/// Gets the <see cref="HonestWorkerRunner" /> results.
/// </summary>
/// <param name="runner">
/// The runner.
/// </param>
/// <returns>
/// The dictionary of results.
/// </returns>
public static Dictionary <string, object> GetHonestWorkerTrainingResults(ModelRunnerBase runner)
{
var result = GetModelTrainingResults(runner);
var abilityHistogram = PlotData.GetWorkerAbilities((HonestWorkerRunner)runner);
result["AbilityHistogram"] = abilityHistogram;
return(result);
}
/// <summary>
/// Gets <see cref="RunnerBase" /> results.
/// </summary>
/// <param name="runner">
/// The runner.
/// </param>
/// <returns>
/// The dictionary of results.
/// </returns>
public static Dictionary <string, object> GetTrainingResults(RunnerBase runner)
{
var result = new Dictionary <string, object>();
var confusionMatrix = PlotData.GetConfusionMatrix(runner.ConfusionMatrix, runner.DataMapping,
PlotData.ConfusionMatrixRowLabel, PlotData.ConfusionMatrixColLabel2);
result["ConfusionMatrix"] = confusionMatrix;
var confusionMatrixPercentage = PlotData.GetConfusionMatrix(runner.ConfusionMatrix, runner.DataMapping,
PlotData.ConfusionMatrixRowLabel, PlotData.ConfusionMatrixColLabel2, true);
result["ConfusionMatrixPercentage"] = confusionMatrixPercentage;
var tweetMatrix = PlotData.GetTweetMatrix(runner.TweetMatrix, runner.DataMapping,
PlotData.ConfusionMatrixRowLabel, "Inferred True");
result["Accuracy"] = runner.Accuracy;
result["AverageRecall"] = runner.AverageRecall;
return(result);
}
/// <summary>
/// Gets the validation results.
/// </summary>
/// <param name="runner">
/// The runner.
/// </param>
/// <param name="validationMetrics">
/// The validation metrics.
/// </param>
/// <returns>
/// The validation results.
/// </returns>
public static Dictionary <string, object> GetValidationResults(ModelRunnerBase runner,
Dictionary <string, object> validationMetrics)
{
var results = new Dictionary <string, object>();
foreach (var kvp in validationMetrics)
{
var key = kvp.Key;
var value = kvp.Value;
if (kvp.Key == Metric.ConfusionMatrix.ToString())
{
value = PlotData.GetConfusionMatrix((double[, ])kvp.Value, runner.DataMapping,
PlotData.ConfusionMatrixRowLabel, PlotData.ConfusionMatrixColLabel2);
var valuePerc = PlotData.GetConfusionMatrix((double[, ])kvp.Value, runner.DataMapping,
PlotData.ConfusionMatrixRowLabel, PlotData.ConfusionMatrixColLabel2, true);
results[key + "Percentage"] = valuePerc;
}
results[key] = value;
}
return(results);
}
/// <summary>
/// Gets the worker metrics for a data set
/// </summary>
/// <param name="data">
/// The data.
/// </param>
/// <param name="runner">
/// The runner.
/// </param>
/// <param name="maxNumberWorkers">
/// The maximum number of workers for which to get metrics.
/// </param>
/// <returns>
/// The metrics.
/// </returns>
public static Dictionary <string, Dictionary <string, object> > GetWorkerMetrics(CrowdData data,
RunnerBase runner, int maxNumberWorkers = 20)
{
var metricsForWorkers = new List <Metric> {
Metric.Count, Metric.Accuracy, Metric.ConfusionMatrix
};
var result = new Dictionary <string, Dictionary <string, object> >();
var mapping = runner != null ? runner.DataMapping : new CrowdDataMapping(data, LabelValuesToString);
foreach (var metric in metricsForWorkers)
{
result[metric.ToString()] = new Dictionary <string, object>();
}
var confusionMatrixKey = Metric.ConfusionMatrix.ToString();
var confusionMatrixPercentageKey = $"{confusionMatrixKey}Percentage";
result[confusionMatrixPercentageKey] = new Dictionary <string, object>();
var labelsGroupedByWorker = data.CrowdLabels.GroupBy(cd => cd.WorkerId);
var trueLabels = runner?.Predictions;
const string RowLabel = PlotData.ConfusionMatrixRowLabel;
const string ColumnLabel = PlotData.ConfusionMatrixColLabel;
foreach (var worker in labelsGroupedByWorker)
{
var workerId = worker.Key;
var workerLabels = worker.Distinct(CrowdData.WorkerTweetEqualityComparer.Instance)
.ToDictionary(lab => lab.TweetId, lab => lab.WorkerLabel);
var workerMetrics = GetMetrics(mapping, workerLabels, trueLabels);
foreach (var metric in metricsForWorkers)
{
if (metric == Metric.ConfusionMatrix)
{
var mat = workerMetrics[metric];
var confMat = PlotData.GetConfusionMatrix((double[, ])mat, mapping, RowLabel, ColumnLabel);
result[confusionMatrixKey][workerId] = confMat;
var matPerc =
PlotData.GetConfusionMatrix((double[, ])mat, mapping, RowLabel, ColumnLabel, true);
result[confusionMatrixPercentageKey][workerId] = matPerc;
}
else
{
result[metric.ToString()][workerId] = workerMetrics[metric];
}
}
}
// Limit the confusion matrices to the more prominent workers.
var prominentWorkers = new HashSet <string>(result[Metric.Count.ToString()]
.OrderByDescending(kvp => (int)kvp.Value).Take(maxNumberWorkers).Select(kvp => kvp.Key));
result[confusionMatrixKey] = result[confusionMatrixKey].Where(kvp1 => prominentWorkers.Contains(kvp1.Key))
.ToDictionary(kvp1 => kvp1.Key, kvp1 => kvp1.Value);
result[confusionMatrixPercentageKey] = result[confusionMatrixPercentageKey]
.Where(kvp1 => prominentWorkers.Contains(kvp1.Key))
.ToDictionary(kvp1 => kvp1.Key, kvp1 => kvp1.Value);
return(result);
}