 Visualizza su TensorFlow.org |  Esegui in Google Colab |  Visualizza la fonte su GitHub |  Scarica taccuino |
In questo tutorial, stiamo andando a formare lo stesso modello di recupero come abbiamo fatto nel recupero di base tutorial, ma con la strategia di distribuzione.
Stava andando a:
- Ottieni i nostri dati e dividili in un set di allenamento e test.
- Configura due GPU virtuali e TensorFlow MirroredStrategy.
- Implementa un modello di recupero utilizzando MirroredStrategy.
- Adattalo a MirrorredStrategy e valutalo.
Importazioni
Per prima cosa togliamo di mezzo le nostre importazioni.
pip install -q tensorflow-recommenderspip install -q --upgrade tensorflow-datasets
import os
import pprint
import tempfile
from typing import Dict, Text
import numpy as np
import tensorflow as tf
import tensorflow_datasets as tfds
import tensorflow_recommenders as tfrs
Preparazione del set di dati
Prepariamo il set di dati esattamente nello stesso modo in cui facciamo nel recupero di base tutorial.
# Ratings data.
ratings = tfds.load("movielens/100k-ratings", split="train")
# Features of all the available movies.
movies = tfds.load("movielens/100k-movies", split="train")
for x in ratings.take(1).as_numpy_iterator():
pprint.pprint(x)
for x in movies.take(1).as_numpy_iterator():
pprint.pprint(x)
ratings = ratings.map(lambda x: {
"movie_title": x["movie_title"],
"user_id": x["user_id"],
})
movies = movies.map(lambda x: x["movie_title"])
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)
movie_titles = movies.batch(1_000)
user_ids = ratings.batch(1_000_000).map(lambda x: x["user_id"])
unique_movie_titles = np.unique(np.concatenate(list(movie_titles)))
unique_user_ids = np.unique(np.concatenate(list(user_ids)))
unique_movie_titles[:10]
{'bucketized_user_age': 45.0,
'movie_genres': array([7]),
'movie_id': b'357',
'movie_title': b"One Flew Over the Cuckoo's Nest (1975)",
'raw_user_age': 46.0,
'timestamp': 879024327,
'user_gender': True,
'user_id': b'138',
'user_occupation_label': 4,
'user_occupation_text': b'doctor',
'user_rating': 4.0,
'user_zip_code': b'53211'}
2021-10-14 11:16:44.748468: W tensorflow/core/kernels/data/cache_dataset_ops.cc:768] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead.
{'movie_genres': array([4]),
'movie_id': b'1681',
'movie_title': b'You So Crazy (1994)'}
2021-10-14 11:16:45.396856: W tensorflow/core/kernels/data/cache_dataset_ops.cc:768] The calling iterator did not fully read the dataset being cached. In order to avoid unexpected truncation of the dataset, the partially cached contents of the dataset will be discarded. This can happen if you have an input pipeline similar to `dataset.cache().take(k).repeat()`. You should use `dataset.take(k).cache().repeat()` instead.
array([b"'Til There Was You (1997)", b'1-900 (1994)',
b'101 Dalmatians (1996)', b'12 Angry Men (1957)', b'187 (1997)',
b'2 Days in the Valley (1996)',
b'20,000 Leagues Under the Sea (1954)',
b'2001: A Space Odyssey (1968)',
b'3 Ninjas: High Noon At Mega Mountain (1998)',
b'39 Steps, The (1935)'], dtype=object)
Configura due GPU virtuali
Se non hai aggiunto acceleratori GPU al tuo Colab, disconnetti il runtime di Colab e fallo ora. Abbiamo bisogno della GPU per eseguire il codice seguente:
gpus = tf.config.list_physical_devices("GPU")
if gpus:
# Create 2 virtual GPUs with 1GB memory each
try:
tf.config.set_logical_device_configuration(
gpus[0],
[tf.config.LogicalDeviceConfiguration(memory_limit=1024),
tf.config.LogicalDeviceConfiguration(memory_limit=1024)])
logical_gpus = tf.config.list_logical_devices("GPU")
print(len(gpus), "Physical GPU,", len(logical_gpus), "Logical GPUs")
except RuntimeError as e:
# Virtual devices must be set before GPUs have been initialized
print(e)
strategy = tf.distribute.MirroredStrategy()
Virtual devices cannot be modified after being initialized
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0',)
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0',)
Implementazione di un modello
Implementiamo le user_model, movie_model, metriche e compito nello stesso modo come facciamo nel recupero di base tutorial, ma noi li avvolgiamo nella dotazione strategia di distribuzione:
embedding_dimension = 32
with strategy.scope():
user_model = tf.keras.Sequential([
tf.keras.layers.StringLookup(
vocabulary=unique_user_ids, mask_token=None),
# We add an additional embedding to account for unknown tokens.
tf.keras.layers.Embedding(len(unique_user_ids) + 1, embedding_dimension)
])
movie_model = tf.keras.Sequential([
tf.keras.layers.StringLookup(
vocabulary=unique_movie_titles, mask_token=None),
tf.keras.layers.Embedding(len(unique_movie_titles) + 1, embedding_dimension)
])
metrics = tfrs.metrics.FactorizedTopK(
candidates=movies.batch(128).map(movie_model)
)
task = tfrs.tasks.Retrieval(
metrics=metrics
)
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
Ora possiamo mettere tutto insieme in un modello. Questa è esattamente la stessa come nel recupero di base tutorial.
class MovielensModel(tfrs.Model):
def __init__(self, user_model, movie_model):
super().__init__()
self.movie_model: tf.keras.Model = movie_model
self.user_model: tf.keras.Model = user_model
self.task: tf.keras.layers.Layer = task
def compute_loss(self, features: Dict[Text, tf.Tensor], training=False) -> tf.Tensor:
# We pick out the user features and pass them into the user model.
user_embeddings = self.user_model(features["user_id"])
# And pick out the movie features and pass them into the movie model,
# getting embeddings back.
positive_movie_embeddings = self.movie_model(features["movie_title"])
# The task computes the loss and the metrics.
return self.task(user_embeddings, positive_movie_embeddings)
Adattamento e valutazione
Ora istanziamo e compiliamo il modello all'interno dell'ambito della strategia di distribuzione.
Si noti che stiamo usando Adam ottimizzatore qui invece di Adagrad come nel recupero di base esercitazione poiché Adagrad non è supportato qui.
with strategy.scope():
model = MovielensModel(user_model, movie_model)
model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.1))
Quindi mescola, raggruppa e memorizza nella cache i dati di formazione e valutazione.
cached_train = train.shuffle(100_000).batch(8192).cache()
cached_test = test.batch(4096).cache()
Quindi addestrare il modello:
model.fit(cached_train, epochs=3)
2021-10-14 11:16:50.692190: W tensorflow/core/grappler/optimizers/data/auto_shard.cc:461] The `assert_cardinality` transformation is currently not handled by the auto-shard rewrite and will be removed. Epoch 1/3 10/10 [==============================] - 8s 328ms/step - factorized_top_k/top_1_categorical_accuracy: 5.0000e-05 - factorized_top_k/top_5_categorical_accuracy: 8.2500e-04 - factorized_top_k/top_10_categorical_accuracy: 0.0025 - factorized_top_k/top_50_categorical_accuracy: 0.0220 - factorized_top_k/top_100_categorical_accuracy: 0.0537 - loss: 70189.8047 - regularization_loss: 0.0000e+00 - total_loss: 70189.8047 Epoch 2/3 10/10 [==============================] - 3s 329ms/step - factorized_top_k/top_1_categorical_accuracy: 3.3750e-04 - factorized_top_k/top_5_categorical_accuracy: 0.0113 - factorized_top_k/top_10_categorical_accuracy: 0.0251 - factorized_top_k/top_50_categorical_accuracy: 0.1268 - factorized_top_k/top_100_categorical_accuracy: 0.2325 - loss: 66736.4560 - regularization_loss: 0.0000e+00 - total_loss: 66736.4560 Epoch 3/3 10/10 [==============================] - 3s 332ms/step - factorized_top_k/top_1_categorical_accuracy: 0.0012 - factorized_top_k/top_5_categorical_accuracy: 0.0198 - factorized_top_k/top_10_categorical_accuracy: 0.0417 - factorized_top_k/top_50_categorical_accuracy: 0.1834 - factorized_top_k/top_100_categorical_accuracy: 0.3138 - loss: 64871.2997 - regularization_loss: 0.0000e+00 - total_loss: 64871.2997 <keras.callbacks.History at 0x7fb74c479190>
Puoi vedere dal registro di formazione che TFRS utilizza entrambe le GPU virtuali.
Infine, possiamo valutare il nostro modello sul test set:
model.evaluate(cached_test, return_dict=True)
2021-10-14 11:17:05.371963: W tensorflow/core/grappler/optimizers/data/auto_shard.cc:461] The `assert_cardinality` transformation is currently not handled by the auto-shard rewrite and will be removed.
5/5 [==============================] - 4s 193ms/step - factorized_top_k/top_1_categorical_accuracy: 5.0000e-05 - factorized_top_k/top_5_categorical_accuracy: 0.0013 - factorized_top_k/top_10_categorical_accuracy: 0.0043 - factorized_top_k/top_50_categorical_accuracy: 0.0639 - factorized_top_k/top_100_categorical_accuracy: 0.1531 - loss: 32404.8092 - regularization_loss: 0.0000e+00 - total_loss: 32404.8092
{'factorized_top_k/top_1_categorical_accuracy': 4.999999873689376e-05,
'factorized_top_k/top_5_categorical_accuracy': 0.0013000000035390258,
'factorized_top_k/top_10_categorical_accuracy': 0.00430000014603138,
'factorized_top_k/top_50_categorical_accuracy': 0.06385000050067902,
'factorized_top_k/top_100_categorical_accuracy': 0.1530500054359436,
'loss': 29363.98046875,
'regularization_loss': 0,
'total_loss': 29363.98046875}
Questo conclude il recupero con l'esercitazione sulla strategia di distribuzione.