protected override PolicyDecision <ActionProbability[]> MapContext(VowpalWabbit vw, string context)
{
using (var vwJson = new VowpalWabbitJsonSerializer(vw))
using (VowpalWabbitExampleCollection vwExample = vwJson.ParseAndCreate(context))
{
if (this.developmentMode)
{
Trace.TraceInformation("Example Context: '{0}'", vwExample.VowpalWabbitString);
}
var vwPredictions = vwExample.Predict(VowpalWabbitPredictionType.ActionProbabilities);
// VW multi-label predictions are 0-based
var ap = vwPredictions
.Select(a =>
new ActionProbability
{
Action = (int)(a.Action + 1),
Probability = a.Score
})
.ToArray();
var state = new VWState {
ModelId = vw.Native.ID
};
return(PolicyDecision.Create(ap, state));
}
}
public void ValidateExample(VowpalWabbitExampleCollection example, Context ctx)
{
using (var vw = new VowpalWabbit(new VowpalWabbitSettings {
EnableStringExampleGeneration = true
}))
using (var validator = new VowpalWabbitExampleJsonValidator())
{
var singleExample = (VowpalWabbitSingleLineExampleCollection)example;
validator.Validate(ctx.VW, example);
}
}
/// <summary>
/// Determines the action to take for a given context.
/// This implementation should be thread-safe if multithreading is needed.
/// </summary>
/// <param name="vw">The Vowpal Wabbit instance to use.</param>
/// <param name="context">A user-defined context for the decision.</param>
/// <returns>A decision tuple containing the index of the action to take (1-based), and the Id of the model or policy used to make the decision.
/// Can be null if the Policy is not ready yet (e.g. model not loaded).</returns>
protected override PolicyDecision <int> MapContext(VowpalWabbit vw, string context)
{
using (var vwJson = new VowpalWabbitJsonSerializer(vw))
using (VowpalWabbitExampleCollection vwExample = vwJson.ParseAndCreate(context))
{
var action = (int)vwExample.Predict(VowpalWabbitPredictionType.CostSensitive);
var state = new VWState {
ModelId = vw.ID
};
return(PolicyDecision.Create(action, state));
}
}
/// <summary>
/// Determines the action to take for a given context.
/// This implementation should be thread-safe if multithreading is needed.
/// </summary>
/// <param name="vw">The Vowpal Wabbit instance to use.</param>
/// <param name="context">A user-defined context for the decision.</param>
/// <returns>A decision tuple containing the index of the action to take (1-based), and the Id of the model or policy used to make the decision.
/// Can be null if the Policy is not ready yet (e.g. model not loaded).</returns>
protected override PolicyDecision <int[]> MapContext(VowpalWabbit vw, string context)
{
using (var vwJson = new VowpalWabbitJsonSerializer(vw))
using (VowpalWabbitExampleCollection vwExample = vwJson.ParseAndCreate(context))
{
ActionScore[] vwMultilabelPredictions = vwExample.Predict(VowpalWabbitPredictionType.ActionProbabilities);
// VW multi-label predictions are 0-based
var actions = vwMultilabelPredictions.Select(a => (int)a.Action + 1).ToArray();
var state = new VWState {
ModelId = vw.ID
};
return(PolicyDecision.Create(actions, state));
}
}
private void updateModelMaybe()
{
if (sinceLastUpdate >= ModelUpdateInterval)
{
// Locking at this level ensures a batch of events is processed completely before
// the next batch (finer locking would allow interleaving, violating timeorder
lock (this.vwLock)
{
// Exit gracefully if the object has been disposed
if (vwDisposed)
{
return;
}
foreach (var dp in log.FlushCompleteEvents())
{
uint action = (uint)((int[])dp.InteractData.Value)[0];
var label = new ContextualBanditLabel(action, -dp.Reward, ((GenericTopSlotExplorerState)dp.InteractData.ExplorerState).Probabilities[0]);
// String (json) contexts need to be handled specially, since the C# interface
// does not currently handle the CB label properly
if (typeof(TContext) == typeof(string))
{
// Manually insert the CB label fields into the context
string labelStr = string.Format(CultureInfo.InvariantCulture, "\"_label_Action\":{0},\"_label_Cost\":{1},\"_label_Probability\":{2},\"_labelIndex\":{3},",
label.Action, label.Cost, label.Probability, label.Action - 1);
string context = ((string)dp.InteractData.Context).Insert(1, labelStr);
using (var vwSerializer = new VowpalWabbitJsonSerializer(vwJson.Native))
using (VowpalWabbitExampleCollection vwExample = vwSerializer.ParseAndCreate(context))
{
vwExample.Learn();
}
}
else
{
vw.Learn((TContext)dp.InteractData.Context, label, index: (int)label.Action - 1);
}
}
using (MemoryStream currModel = new MemoryStream())
{
VowpalWabbit vwNative = (typeof(TContext) == typeof(string)) ? vwJson.Native : vw.Native;
vwNative.SaveModel(currModel);
currModel.Position = 0;
this.UpdateModel(currModel);
sinceLastUpdate = 0;
}
}
}
}
public void Validate(string line, VowpalWabbitExampleCollection example, IVowpalWabbitLabelComparator labelComparator = null, ILabel label = null)
{
Assert.IsNotNull(example);
var jsonExample = example as VowpalWabbitSingleLineExampleCollection;
Assert.IsNotNull(jsonExample);
using (var strExample = this.vw.ParseLine(line))
using (var strJsonExample = this.vw.ParseLine(jsonExample.Example.VowpalWabbitString))
{
var diff = strExample.Diff(this.vw, jsonExample.Example, labelComparator);
Assert.IsNull(diff, diff + " generated string: '" + jsonExample.VowpalWabbitString + "'");
diff = strExample.Diff(this.vw, strJsonExample, labelComparator);
Assert.IsNull(diff, diff);
}
}
public void Validate(string line, VowpalWabbitExampleCollection example, IVowpalWabbitLabelComparator labelComparator = null, ILabel label = null)
{
Assert.IsNotNull(example);
var jsonExample = example as VowpalWabbitSingleLineExampleCollection;
Assert.IsNotNull(jsonExample);
using (var strExample = this.vw.ParseLine(line))
using (var strJsonExample = this.vw.ParseLine(jsonExample.Example.VowpalWabbitString))
{
var diff = strExample.Diff(this.vw, jsonExample.Example, labelComparator);
Assert.IsNull(diff, diff + " generated string: '" + jsonExample.VowpalWabbitString + "'");
diff = strExample.Diff(this.vw, strJsonExample, labelComparator);
Assert.IsNull(diff, diff);
}
}
