什么是特性交叉，为什么它们很重要?想象一下，我们正在建立一个向客户出售搅拌机的推荐系统。然后，客户过去的购买历史，如purchased_bananas和purchased_cooking_books，或地理特征，都是单一特征。如果一个人同时购买了香蕉和烹饪书籍，那么这个顾客更有可能点击推荐的搅拌机。purchased_bananas和purchased_cooking_books的组合被称为功能交叉，它提供了单个功能之外的额外交互信息。
学习特性交叉的挑战是什么?在web规模的应用程序中，数据大多是分类的，导致了大而稀疏的特征空间。在这种情况下，识别有效的特征交叉通常需要手工特征工程或穷举搜索。传统的前馈多层感知器模型是通用函数逼近器;然而，它们甚至不能有效地逼近二阶或三阶特征交叉

import pprint

%matplotlib inline
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable

import numpy as np
import tensorflow as tf
import tensorflow_datasets as tfds

import tensorflow_recommenders as tfrs

我们首先建立了一个以均方误差为损失的统一模型类

class Model(tfrs.Model):

  def __init__(self, model):
    super().__init__()
    self._model = model
    self._logit_layer = tf.keras.layers.Dense(1)

    self.task = tfrs.tasks.Ranking(
      loss=tf.keras.losses.MeanSquaredError(),
      metrics=[
        tf.keras.metrics.RootMeanSquaredError("RMSE")
      ]
    )

  def call(self, x):
    x = self._model(x)
    return self._logit_layer(x)

  def compute_loss(self, features, training=False):
    x, labels = features
    scores = self(x)

    return self.task(
        labels=labels,
        predictions=scores,
    )

指定cross network 和 ReLU-based DNN

crossnet = Model(tfrs.layers.dcn.Cross())
deepnet = Model(
    tf.keras.Sequential([
      tf.keras.layers.Dense(512, activation="relu"),
      tf.keras.layers.Dense(256, activation="relu"),
      tf.keras.layers.Dense(128, activation="relu")
    ])
)

数据集

ratings = tfds.load("movie_lens/100k-ratings", split="train")
ratings = ratings.map(lambda x: {
    "movie_id": x["movie_id"],
    "user_id": x["user_id"],
    "user_rating": x["user_rating"],
    "user_gender": int(x["user_gender"]),
    "user_zip_code": x["user_zip_code"],
    "user_occupation_text": x["user_occupation_text"],
    "bucketized_user_age": int(x["bucketized_user_age"]),
})

随机划分训练集和测试集

tf.random.set_seed(42)
shuffled = ratings.shuffle(100_000, seed=42, reshuffle_each_iteration=False)

train = shuffled.take(80_000)
test = shuffled.skip(80_000).take(20_000)

特征词汇表

feature_names = ["movie_id", "user_id", "user_gender", "user_zip_code",
                 "user_occupation_text", "bucketized_user_age"]

vocabularies = {}

for feature_name in feature_names:
  vocab = ratings.batch(1_000_000).map(lambda x: x[feature_name])
  vocabularies[feature_name] = np.unique(np.concatenate(list(vocab)))

模型结构：

class DCN(tfrs.Model):

  def __init__(self, use_cross_layer, deep_layer_sizes, projection_dim=None):
    super().__init__()

    self.embedding_dimension = 32

    str_features = ["movie_id", "user_id", "user_zip_code",
                    "user_occupation_text"]
    int_features = ["user_gender", "bucketized_user_age"]

    self._all_features = str_features + int_features
    self._embeddings = {}

    # Compute embeddings for string features.
    for feature_name in str_features:
      vocabulary = vocabularies[feature_name]
      self._embeddings[feature_name] = tf.keras.Sequential(
          [tf.keras.layers.StringLookup(
              vocabulary=vocabulary, mask_token=None),
           tf.keras.layers.Embedding(len(vocabulary) + 1,
                                     self.embedding_dimension)
    ])

    # Compute embeddings for int features.
    for feature_name in int_features:
      vocabulary = vocabularies[feature_name]
      self._embeddings[feature_name] = tf.keras.Sequential(
          [tf.keras.layers.IntegerLookup(
              vocabulary=vocabulary, mask_value=None),
           tf.keras.layers.Embedding(len(vocabulary) + 1,
                                     self.embedding_dimension)
    ])

    if use_cross_layer:
      self._cross_layer = tfrs.layers.dcn.Cross(
          projection_dim=projection_dim,
          kernel_initializer="glorot_uniform")
    else:
      self._cross_layer = None

    self._deep_layers = [tf.keras.layers.Dense(layer_size, activation="relu")
      for layer_size in deep_layer_sizes]

    self._logit_layer = tf.keras.layers.Dense(1)

    self.task = tfrs.tasks.Ranking(
      loss=tf.keras.losses.MeanSquaredError(),
      metrics=[tf.keras.metrics.RootMeanSquaredError("RMSE")]
    )

  def call(self, features):
    # Concatenate embeddings
    embeddings = []
    for feature_name in self._all_features:
      embedding_fn = self._embeddings[feature_name]
      embeddings.append(embedding_fn(features[feature_name]))

    x = tf.concat(embeddings, axis=1)

    # Build Cross Network
    if self._cross_layer is not None:
      x = self._cross_layer(x)

    # Build Deep Network
    for deep_layer in self._deep_layers:
      x = deep_layer(x)

    return self._logit_layer(x)

  def compute_loss(self, features, training=False):
    labels = features.pop("user_rating")
    scores = self(features)
    return self.task(
        labels=labels,
        predictions=scores,
    )

训练、测试集

cached_train = train.shuffle(100_000).batch(8192).cache()
cached_test = test.batch(4096).cache()

让我们定义一个函数，它多次运行一个模型，并从多次运行中返回模型的RMSE平均值和标准偏差。

def run_models(use_cross_layer, deep_layer_sizes, projection_dim=None, num_runs=5):
  models = []
  rmses = []

  for i in range(num_runs):
    model = DCN(use_cross_layer=use_cross_layer,
                deep_layer_sizes=deep_layer_sizes,
                projection_dim=projection_dim)
    model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate))
    models.append(model)

    model.fit(cached_train, epochs=epochs, verbose=False)
    metrics = model.evaluate(cached_test, return_dict=True)
    rmses.append(metrics["RMSE"])

  mean, stdv = np.average(rmses), np.std(rmses)

  return {"model": models, "mean": mean, "stdv": stdv}

设置超参数

epochs = 8
learning_rate = 0.01

我们首先训练一个具有堆叠结构的DCN模型，也就是说，输入被馈送到一个交叉网络，然后是一个深网络。

dcn_result = run_models(use_cross_layer=True,
                        deep_layer_sizes=[192, 192])