public void Train(MalTrainingData trainingData)
{
IBasicTrainingData <IBasicInputForUser> basicTrainingData =
trainingData.AsBasicTrainingData(MinEpisodesToCountIncomplete, UseDropped);
m_recommender.Train(basicTrainingData);
}
private void SetAverages(MalTrainingData trainingData)
{
m_itemAverages.Clear();
Dictionary <int, float> scoreSumByItem = new Dictionary <int, float>();
Dictionary <int, int> numScoresByItem = new Dictionary <int, int>();
IBasicTrainingData <IBasicInputForUser> basicTrainingData =
trainingData.AsBasicTrainingData(m_minEpisodesToClassifyIncomplete, includeDropped: true);
foreach (int userId in basicTrainingData.Users.Keys)
{
foreach (KeyValuePair <int, float> itemIdRatingPair in basicTrainingData.Users[userId].Ratings)
{
int itemId = itemIdRatingPair.Key;
float rating = itemIdRatingPair.Value;
if (!scoreSumByItem.ContainsKey(itemId))
{
scoreSumByItem[itemId] = 0;
numScoresByItem[itemId] = 0;
}
scoreSumByItem[itemId] += rating;
numScoresByItem[itemId]++;
}
}
foreach (int itemId in scoreSumByItem.Keys)
{
float averageScore = scoreSumByItem[itemId] / numScoresByItem[itemId];
m_itemAverages[itemId] = averageScore;
}
}
public void Train(MalTrainingData trainingData)
{
IBasicTrainingData <IBasicInputForUser> basicTrainingData =
trainingData.AsBasicTrainingData(MinEpisodesToCountIncomplete, UseDropped);
IBasicTrainingData <IBasicInputForUser> filteredTrainingData = FilterHelpers.RemoveItemsWithFewUsers(basicTrainingData, MinUsersToCountAnime);
m_recommender.Train(filteredTrainingData);
}
public void Train(MalTrainingData trainingData)
{
m_userCountByAnime = new Dictionary <int, int>();
IBasicTrainingData <IBasicInputForUser> basicTrainingData =
trainingData.AsBasicTrainingData(m_minEpisodesToCountIncomplete, m_useDropped);
foreach (int userId in basicTrainingData.Users.Keys)
{
foreach (int itemId in basicTrainingData.Users[userId].Ratings.Keys)
{
if (!m_userCountByAnime.ContainsKey(itemId))
{
m_userCountByAnime[itemId] = 0;
}
m_userCountByAnime[itemId]++;
}
}
// Do not filter out anime with few users at this point. Let the MyMediaLite recommender possibly make use of that data.
m_recommender.Train(basicTrainingData);
}
public void Train(IBasicTrainingData <IBasicInputForUser> trainingData)
{
m_realUserIdToMediaLiteUserId = new Dictionary <int, int>();
m_mediaLiteUserIdToRealUserId = new Dictionary <int, int>();
m_nextMediaLiteUserId = 0;
m_realItemIdToMediaLiteItemId = new Dictionary <int, int>();
m_mediaLiteItemIdToRealItemId = new Dictionary <int, int>();
m_nextMediaLiteItemId = 0;
MyMediaLite.Data.Ratings mediaLiteRatings = new MyMediaLite.Data.Ratings();
foreach (KeyValuePair <int, IBasicInputForUser> userRatingsPair in trainingData.Users)
{
int userId = userRatingsPair.Key;
IBasicInputForUser ratings = userRatingsPair.Value;
m_realUserIdToMediaLiteUserId[userId] = m_nextMediaLiteUserId;
m_mediaLiteUserIdToRealUserId[m_nextMediaLiteUserId] = userId;
m_nextMediaLiteUserId++;
foreach (KeyValuePair <int, float> rating in ratings.Ratings)
{
int itemId = rating.Key;
float score = rating.Value;
if (!m_realItemIdToMediaLiteItemId.ContainsKey(itemId))
{
m_realItemIdToMediaLiteItemId[itemId] = m_nextMediaLiteItemId;
m_mediaLiteItemIdToRealItemId[m_nextMediaLiteItemId] = itemId;
m_nextMediaLiteItemId++;
}
mediaLiteRatings.Add(m_realUserIdToMediaLiteUserId[userId], m_realItemIdToMediaLiteItemId[itemId], score);
}
}
m_recommender.Ratings = mediaLiteRatings;
m_recommender.Train();
}
public void Train(MalTrainingData trainingData)
{
m_userCountByAnime = new Dictionary <int, int>();
IBasicTrainingData <IPositiveFeedbackForUser> basicFeedback = trainingData.AsPositiveFeedback(m_positiveClassifier);
foreach (int userId in basicFeedback.Users.Keys)
{
foreach (int itemId in basicFeedback.Users[userId].Items)
{
if (!m_userCountByAnime.ContainsKey(itemId))
{
m_userCountByAnime[itemId] = 0;
}
m_userCountByAnime[itemId]++;
}
}
// Do not filter out anime with few users at this point. Let the MyMediaLite recommender possibly make use of that data.
m_recommender.Train(basicFeedback);
}
public void Train(IBasicTrainingData <IPositiveFeedbackForUser> trainingData)
{
m_realUserIdToMediaLiteUserId = new Dictionary <int, int>();
m_mediaLiteUserIdToRealUserId = new Dictionary <int, int>();
m_nextMediaLiteUserId = 0;
m_realItemIdToMediaLiteItemId = new Dictionary <int, int>();
m_mediaLiteItemIdToRealItemId = new Dictionary <int, int>();
m_nextMediaLiteItemId = 0;
PosOnlyFeedback <SparseBooleanMatrix> mediaLiteFeedback = new PosOnlyFeedback <SparseBooleanMatrix>();
foreach (KeyValuePair <int, IPositiveFeedbackForUser> userFeedbackPair in trainingData.Users)
{
int userId = userFeedbackPair.Key;
IPositiveFeedbackForUser feedback = userFeedbackPair.Value;
m_realUserIdToMediaLiteUserId[userId] = m_nextMediaLiteUserId;
m_mediaLiteUserIdToRealUserId[m_nextMediaLiteUserId] = userId;
m_nextMediaLiteUserId++;
foreach (int itemId in feedback.Items)
{
if (!m_realItemIdToMediaLiteItemId.ContainsKey(itemId))
{
m_realItemIdToMediaLiteItemId[itemId] = m_nextMediaLiteItemId;
m_mediaLiteItemIdToRealItemId[m_nextMediaLiteItemId] = itemId;
m_nextMediaLiteItemId++;
}
mediaLiteFeedback.Add(m_realUserIdToMediaLiteUserId[userId], m_realItemIdToMediaLiteItemId[itemId]);
}
}
m_recommender.Feedback = mediaLiteFeedback;
m_recommender.Train();
}
public static IBasicTrainingData<IBasicInputForUser> RemoveItemsWithFewUsers(IBasicTrainingData<IBasicInputForUser> trainingData, int minimumUsers)
{
Dictionary<int, int> itemRatingCountByItemId = new Dictionary<int, int>();
foreach (int userId in trainingData.Users.Keys)
{
foreach (int animeId in trainingData.Users[userId].Ratings.Keys)
{
if (!itemRatingCountByItemId.ContainsKey(animeId))
{
itemRatingCountByItemId[animeId] = 0;
}
itemRatingCountByItemId[animeId]++;
}
}
foreach (int userId in trainingData.Users.Keys)
{
List<int> animeIdsToRemove = new List<int>();
foreach (int animeId in trainingData.Users[userId].Ratings.Keys)
{
if (itemRatingCountByItemId[animeId] < minimumUsers)
{
animeIdsToRemove.Add(animeId);
}
}
foreach (int animeIdToRemove in animeIdsToRemove)
{
trainingData.Users[userId].Ratings.Remove(animeIdToRemove);
}
}
return trainingData;
}
public static IBasicTrainingData <IBasicInputForUser> RemoveItemsWithFewUsers(IBasicTrainingData <IBasicInputForUser> trainingData, int minimumUsers)
{
Dictionary <int, int> itemRatingCountByItemId = new Dictionary <int, int>();
foreach (int userId in trainingData.Users.Keys)
{
foreach (int animeId in trainingData.Users[userId].Ratings.Keys)
{
if (!itemRatingCountByItemId.ContainsKey(animeId))
{
itemRatingCountByItemId[animeId] = 0;
}
itemRatingCountByItemId[animeId]++;
}
}
foreach (int userId in trainingData.Users.Keys)
{
List <int> animeIdsToRemove = new List <int>();
foreach (int animeId in trainingData.Users[userId].Ratings.Keys)
{
if (itemRatingCountByItemId[animeId] < minimumUsers)
{
animeIdsToRemove.Add(animeId);
}
}
foreach (int animeIdToRemove in animeIdsToRemove)
{
trainingData.Users[userId].Ratings.Remove(animeIdToRemove);
}
}
return(trainingData);
}
