public CrossFoldLearner copy() { CrossFoldLearner r = new CrossFoldLearner(models.size(), numCategories(), numFeatures, prior); r.models.clear(); for (OnlineLogisticRegression model : models) { model.close(); OnlineLogisticRegression newModel = new OnlineLogisticRegression(model.numCategories(), model.numFeatures(), model.prior); newModel.copyFrom(model); r.models.add(newModel); } return r; }
public CrossFoldLearner copy() { CrossFoldLearner r = new CrossFoldLearner(models.size(), numCategories(), numFeatures, prior); r.models.clear(); for (OnlineLogisticRegression model : models) { model.close(); OnlineLogisticRegression newModel = new OnlineLogisticRegression(model.numCategories(), model.numFeatures(), model.prior); newModel.copyFrom(model); r.models.add(newModel); } return r; }
public List<ScoredItem> topItems(User u, int limit) { Vector userVector = new RandomAccessSparseVector(model.numFeatures()); encoder.addUserFeatures(u, userVector); double userScore = model.classifyScalarNoLink(userVector); PriorityQueue<ScoredItem> r = new PriorityQueue<ScoredItem>(); for (Item item : items) { Double itemScore = itemCache.get(item); if (itemScore == null) { Vector v = new RandomAccessSparseVector(model.numFeatures()); encoder.addItemFeatures(item, v); itemScore = model.classifyScalarNoLink(v); itemCache.put(item, itemScore); } long code = encoder.interactionHash(u, item); Double interactionScore = interactionCache.get(code); if (interactionScore == null) { Vector v = new RandomAccessSparseVector(model.numFeatures()); encoder.addInteractions(u, item, v); interactionScore = model.classifyScalarNoLink(v); interactionCache.put(code, interactionScore); } double score = userScore + itemScore + interactionScore; r.add(new ScoredItem(score, item)); while (r.size() > limit) { r.poll(); } } return Lists.newArrayList(r); }
public CrossFoldLearner copy() { CrossFoldLearner r = new CrossFoldLearner(models.size(), numCategories(), numFeatures, prior); r.models.clear(); for (OnlineLogisticRegression model : models) { model.close(); OnlineLogisticRegression newModel = new OnlineLogisticRegression(model.numCategories(), model.numFeatures(), model.prior); newModel.copyFrom(model); r.models.add(newModel); } return r; }
public OnlineLogisticRegression copy() { close(); OnlineLogisticRegression r = new OnlineLogisticRegression(numCategories(), numFeatures(), prior); r.copyFrom(this); return r; }
public OnlineLogisticRegression copy() { close(); OnlineLogisticRegression r = new OnlineLogisticRegression(numCategories(), numFeatures(), prior); r.copyFrom(this); return r; }
public OnlineLogisticRegression copy() { close(); OnlineLogisticRegression r = new OnlineLogisticRegression(numCategories(), numFeatures(), prior); r.copyFrom(this); return r; }
System.out.println("no of features = " + learningAlgo.numFeatures()); System.out.println("Probability of cluster 0 = " + (1.0d - r.get(0))); System.out.println("Probability of cluster 1 = " + r.get(0));
System.out.println("no of features = " + learningAlgo.numFeatures()); System.out.println("Probability of cluster 0 = " + r.get(0)); System.out.println("Probability of cluster 1 = " + r.get(1));