package glintlda
import breeze.linalg.{DenseVector, SparseVector, sum}
import glintlda.util.FastRNG
/**
* A Gibbs sample
* @param features The (sequential and ordered) features
* @param topics The assigned topics
*/
class GibbsSample(val features: Array[Int],
val topics: Array[Int]) extends Serializable {
/**
* Returns a dense topic count
* @param nrOfTopics The total number of topics
* @return The dense topic count
*/
def denseCounts(nrOfTopics: Int): DenseVector[Int] = {
val result = DenseVector.zeros[Int](nrOfTopics)
var i = 0
while (i < topics.length) {
result(topics(i)) += 1
i += 1
}
result
}
/**
* Returns a sparse topic count
*
* @param nrOfTopics The total number of topics
* @return The sparse topic count
*/
def sparseCounts(nrOfTopics: Int): SparseVector[Int] = {
val result = SparseVector.zeros[Int](nrOfTopics)
var i = 0
while (i < topics.length) {
result.update(topics(i), result(topics(i)) + 1)
i += 1
}
result
}
}
object GibbsSample {
/**
* Initializes a Gibbs sample with random (uniform) topic assignments
*
* @param sv The sparse vector representing the document
* @param random The random number generator
* @param topics The number of topics
* @return An initialized Gibbs sample with random (uniform) topic assignments
*/
def apply(sv: SparseVector[Int], random: FastRNG, topics: Int): GibbsSample = {
val totalTokens = sum(sv)
val sample = new GibbsSample(new Array[Int](totalTokens), new Array[Int](totalTokens))
var i = 0
var current = 0
while (i < sv.activeSize) {
val index = sv.indexAt(i)
var value = sv.valueAt(i)
while (value > 0) {
sample.features(current) = index
sample.topics(current) = random.nextPositiveInt() % topics
current += 1
value -= 1
}
i += 1
}
sample
}
}
