次の方法で共有


FieldAwareFactorizationMachineModelParameters.GetLatentWeights メソッド

定義

各特徴の潜在的な表現。 1 つの特徴に潜在ベクトルがあり FieldCount 、各潜在ベクトルに値が含まれていることに LatentDimension 注意してください。 f 番目のフィールドでは、ドキュメントhttps://github.com/wschin/fast-ffm/blob/master/fast-ffm.pdf内の j 番目の特徴の潜在ベクトルは、 v_{j, f} latentWeights[j * fieldCount * latentDim + f * latentDim] で始まります。 v_{j, f} の k 番目の要素は、latentWeights[j * fieldCount * latentDim + f * latentDim + k] です。 返される値のサイズは featureCount x fieldCount x latentDim です。

public System.Collections.Generic.IReadOnlyList<float> GetLatentWeights ();
member this.GetLatentWeights : unit -> System.Collections.Generic.IReadOnlyList<single>
Public Function GetLatentWeights () As IReadOnlyList(Of Single)

戻り値

適用対象