public void BinaryRecommendationCustomSerializationRegressionTest()
{
// TODO: We need a static class which implements mapping chainings
Rand.Restart(12347);
var positiveOnlyMapping = new PositiveOnlyDataMapping();
var negativeDataGeneratorMapping = positiveOnlyMapping.WithGeneratedNegativeData(1.0);
using (Stream stream = new MemoryStream())
{
negativeDataGeneratorMapping.SaveForwardCompatibleAsBinary(stream);
stream.Seek(0, SeekOrigin.Begin);
using (var reader = new WrappedBinaryReader(new BinaryReader(stream)))
{
var deserializedMapping =
new NegativeDataGeneratorMapping <PositiveOnlyDataset, Tuple <string, string>, string, string, FeatureProvider, Vector>(reader, positiveOnlyMapping);
var recommender = MatchboxRecommender.Create(deserializedMapping);
recommender.Train(this.positiveOnlyDataset, this.featureProvider); // must not throw
foreach (var instance in positiveOnlyMapping.GetInstances(this.positiveOnlyDataset))
{
var user = positiveOnlyMapping.GetUser(this.positiveOnlyDataset, instance);
var item = positiveOnlyMapping.GetItem(this.positiveOnlyDataset, instance);
Assert.Equal(1, recommender.Predict(user, item, this.featureProvider));
}
}
}
}
/// <summary>
/// Computes the number of original observations, positive
/// ratings, and negative ratings for each user and item
/// </summary>
/// <param name="userRatingCounts">A map from each user to their rating counts.</param>
/// <param name="itemRatingCounts">A map from each item to its rating counts.</param>
private void ComputeRatingCounts(
out IDictionary <string, RatingCounts> userRatingCounts,
out IDictionary <string, RatingCounts> itemRatingCounts)
{
Rand.Restart(12347);
var positiveOnlyMapping = new PositiveOnlyDataMapping();
var negativeDataGeneratorMapping = positiveOnlyMapping.WithGeneratedNegativeData(1.0);
userRatingCounts = new Dictionary <string, RatingCounts>();
itemRatingCounts = new Dictionary <string, RatingCounts>();
// Count the number of positive observations for each user and item.
foreach (var instance in positiveOnlyMapping.GetInstances(this.positiveOnlyDataset))
{
var user = positiveOnlyMapping.GetUser(this.positiveOnlyDataset, instance);
var item = positiveOnlyMapping.GetItem(this.positiveOnlyDataset, instance);
RatingCounts ratingCounts;
if (!userRatingCounts.TryGetValue(user, out ratingCounts))
{
ratingCounts = new RatingCounts();
userRatingCounts.Add(user, ratingCounts);
}
++ratingCounts.Original;
if (!itemRatingCounts.TryGetValue(item, out ratingCounts))
{
ratingCounts = new RatingCounts();
itemRatingCounts.Add(item, ratingCounts);
}
++ratingCounts.Original;
}
// Compute the number of positive and negative ratings generated for each user and item.
foreach (var instance in negativeDataGeneratorMapping.GetInstances(this.positiveOnlyDataset))
{
var user = negativeDataGeneratorMapping.GetUser(this.positiveOnlyDataset, instance);
var item = negativeDataGeneratorMapping.GetItem(this.positiveOnlyDataset, instance);
var rating = negativeDataGeneratorMapping.GetRating(this.positiveOnlyDataset, instance);
if (rating == 1)
{
++userRatingCounts[user].Positive;
++itemRatingCounts[item].Positive;
}
else if (rating == 0)
{
++userRatingCounts[user].Negative;
++itemRatingCounts[item].Negative;
}
}
}
public void BinaryRecommendationRegressionTest()
{
// TODO: We need a static class which implements mapping chainings
Rand.Restart(12347);
var positiveOnlyMapping = new PositiveOnlyDataMapping();
var negativeDataGeneratorMapping = positiveOnlyMapping.WithGeneratedNegativeData(1.0);
var recommender = MatchboxRecommender.Create(negativeDataGeneratorMapping);
recommender.Train(this.positiveOnlyDataset, this.featureProvider); // must not throw
foreach (var instance in positiveOnlyMapping.GetInstances(this.positiveOnlyDataset))
{
var user = positiveOnlyMapping.GetUser(this.positiveOnlyDataset, instance);
var item = positiveOnlyMapping.GetItem(this.positiveOnlyDataset, instance);
Assert.Equal(1, recommender.Predict(user, item, this.featureProvider));
}
}
