Recomendadores do TensorFlow: início rápido

Ver no TensorFlow.org Executar no Google Colab Ver fonte no GitHub Baixar caderno

Neste tutorial, vamos construir um modelo de fatoração matriz simples usando o conjunto de dados MovieLens 100K com TFRS. Podemos usar este modelo para recomendar filmes para um determinado usuário.

Importar TFRS

Primeiro, instale e importe o TFRS:

pip install -q tensorflow-recommenders
pip install -q --upgrade tensorflow-datasets
from typing import Dict, Text

import numpy as np
import tensorflow as tf

import tensorflow_datasets as tfds
import tensorflow_recommenders as tfrs

Leia os dados

# Ratings data.
ratings = tfds.load('movielens/100k-ratings', split="train")
# Features of all the available movies.
movies = tfds.load('movielens/100k-movies', split="train")

# Select the basic features.
ratings = ratings.map(lambda x: {
    "movie_title": x["movie_title"],
    "user_id": x["user_id"]
})
movies = movies.map(lambda x: x["movie_title"])
2021-08-25 11:16:39.656990: E tensorflow/stream_executor/cuda/cuda_driver.cc:271] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detected

Crie vocabulários para converter IDs de usuário e títulos de filmes em índices inteiros para camadas de incorporação:

user_ids_vocabulary = tf.keras.layers.experimental.preprocessing.StringLookup(mask_token=None)
user_ids_vocabulary.adapt(ratings.map(lambda x: x["user_id"]))

movie_titles_vocabulary = tf.keras.layers.experimental.preprocessing.StringLookup(mask_token=None)
movie_titles_vocabulary.adapt(movies)

Defina um modelo

Podemos definir um modelo TFRS por herança de tfrs.Model e implementação do compute_loss método:

class MovieLensModel(tfrs.Model):
  # We derive from a custom base class to help reduce boilerplate. Under the hood,
  # these are still plain Keras Models.

  def __init__(
      self,
      user_model: tf.keras.Model,
      movie_model: tf.keras.Model,
      task: tfrs.tasks.Retrieval):
    super().__init__()

    # Set up user and movie representations.
    self.user_model = user_model
    self.movie_model = movie_model

    # Set up a retrieval task.
    self.task = task

  def compute_loss(self, features: Dict[Text, tf.Tensor], training=False) -> tf.Tensor:
    # Define how the loss is computed.

    user_embeddings = self.user_model(features["user_id"])
    movie_embeddings = self.movie_model(features["movie_title"])

    return self.task(user_embeddings, movie_embeddings)

Defina os dois modelos e a tarefa de recuperação.

# Define user and movie models.
user_model = tf.keras.Sequential([
    user_ids_vocabulary,
    tf.keras.layers.Embedding(user_ids_vocabulary.vocab_size(), 64)
])
movie_model = tf.keras.Sequential([
    movie_titles_vocabulary,
    tf.keras.layers.Embedding(movie_titles_vocabulary.vocab_size(), 64)
])

# Define your objectives.
task = tfrs.tasks.Retrieval(metrics=tfrs.metrics.FactorizedTopK(
    movies.batch(128).map(movie_model)
  )
)
WARNING:tensorflow:vocab_size is deprecated, please use vocabulary_size.
WARNING:tensorflow:vocab_size is deprecated, please use vocabulary_size.
WARNING:tensorflow:vocab_size is deprecated, please use vocabulary_size.
WARNING:tensorflow:vocab_size is deprecated, please use vocabulary_size.

Ajustar e avaliar.

Crie o modelo, treine-o e gere previsões:

# Create a retrieval model.
model = MovieLensModel(user_model, movie_model, task)
model.compile(optimizer=tf.keras.optimizers.Adagrad(0.5))

# Train for 3 epochs.
model.fit(ratings.batch(4096), epochs=3)

# Use brute-force search to set up retrieval using the trained representations.
index = tfrs.layers.factorized_top_k.BruteForce(model.user_model)
index.index_from_dataset(
    movies.batch(100).map(lambda title: (title, model.movie_model(title))))

# Get some recommendations.
_, titles = index(np.array(["42"]))
print(f"Top 3 recommendations for user 42: {titles[0, :3]}")
Epoch 1/3
25/25 [==============================] - 5s 156ms/step - factorized_top_k/top_1_categorical_accuracy: 7.0000e-05 - factorized_top_k/top_5_categorical_accuracy: 0.0016 - factorized_top_k/top_10_categorical_accuracy: 0.0047 - factorized_top_k/top_50_categorical_accuracy: 0.0439 - factorized_top_k/top_100_categorical_accuracy: 0.1000 - loss: 33088.3344 - regularization_loss: 0.0000e+00 - total_loss: 33088.3344
Epoch 2/3
25/25 [==============================] - 4s 155ms/step - factorized_top_k/top_1_categorical_accuracy: 1.1000e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0048 - factorized_top_k/top_10_categorical_accuracy: 0.0140 - factorized_top_k/top_50_categorical_accuracy: 0.1042 - factorized_top_k/top_100_categorical_accuracy: 0.2104 - loss: 31011.7002 - regularization_loss: 0.0000e+00 - total_loss: 31011.7002
Epoch 3/3
25/25 [==============================] - 4s 154ms/step - factorized_top_k/top_1_categorical_accuracy: 3.8000e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0080 - factorized_top_k/top_10_categorical_accuracy: 0.0219 - factorized_top_k/top_50_categorical_accuracy: 0.1440 - factorized_top_k/top_100_categorical_accuracy: 0.2671 - loss: 30417.0654 - regularization_loss: 0.0000e+00 - total_loss: 30417.0654
Top 3 recommendations for user 42: [b'Rent-a-Kid (1995)' b'House Arrest (1996)' b'Mirage (1995)']