Internal change

PiperOrigin-RevId: 263257133

official/recommendation/ncf_input_pipeline.py

@@ -117,10 +117,7 @@ def create_dataset_from_data_producer(producer, params):
   return train_input_dataset, eval_input_dataset
 
 
-def create_ncf_input_data(params,
-                          producer=None,
-                          input_meta_data=None,
-                          strategy=None):
+def create_ncf_input_data(params, producer=None, input_meta_data=None):
   """Creates NCF training/evaluation dataset.
 
   Args:
@@ -131,9 +128,6 @@ def create_ncf_input_data(params,
     input_meta_data: A dictionary of input metadata to be used when reading data
       from tf record files. Must be specified when params["train_input_dataset"]
       is specified.
-    strategy: Distribution strategy used for distributed training. If specified,
-      used to assert that evaluation batch size is correctly a multiple of
-      total number of devices used.
 
   Returns:
     (training dataset, evaluation dataset, train steps per epoch,
@@ -142,17 +136,6 @@ def create_ncf_input_data(params,
   Raises:
     ValueError: If data is being generated online for when using TPU's.
   """
-  # NCF evaluation metric calculation logic assumes that evaluation data
-  # sample size are in multiples of (1 + number of negative samples in
-  # evaluation) for each device. As so, evaluation batch size must be a
-  # multiple of (number of replicas * (1 + number of negative samples)).
-  num_devices = strategy.num_replicas_in_sync if strategy else 1
-  if (params["eval_batch_size"] % (num_devices *
-                                   (1 + rconst.NUM_EVAL_NEGATIVES))):
-    raise ValueError("Evaluation batch size must be divisible by {} "
-                     "times {}".format(num_devices,
-                                       (1 + rconst.NUM_EVAL_NEGATIVES)))
-
   if params["train_dataset_path"]:
     assert params["eval_dataset_path"]
 

official/recommendation/ncf_keras_benchmark.py

@@ -199,7 +199,6 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     self._setup()
     FLAGS.early_stopping = True
     FLAGS.num_gpus = 2
-    FLAGS.eval_batch_size = 160000
     self._run_and_report_benchmark()
 
   def benchmark_2_gpus_ctl_early_stop(self):
@@ -208,7 +207,6 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.keras_use_ctl = True
     FLAGS.early_stopping = True
     FLAGS.num_gpus = 2
-    FLAGS.eval_batch_size = 160000
     self._run_and_report_benchmark()
 
 #############################################
@@ -285,7 +283,6 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.num_gpus = 8
     FLAGS.train_epochs = 17
     FLAGS.batch_size = 1048576
-    FLAGS.eval_batch_size = 160000
     FLAGS.learning_rate = 0.0045
     FLAGS.beta1 = 0.25
     FLAGS.beta2 = 0.5
@@ -298,7 +295,6 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.num_gpus = 8
     FLAGS.train_epochs = 17
     FLAGS.batch_size = 1048576
-    FLAGS.eval_batch_size = 160000
     FLAGS.learning_rate = 0.0045
     FLAGS.beta1 = 0.25
     FLAGS.beta2 = 0.5
@@ -313,7 +309,6 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.num_gpus = 8
     FLAGS.train_epochs = 17
     FLAGS.batch_size = 1048576
-    FLAGS.eval_batch_size = 160000
     FLAGS.learning_rate = 0.0045
     FLAGS.beta1 = 0.25
     FLAGS.beta2 = 0.5
@@ -334,7 +329,6 @@ class NCFKerasSynth(NCFKerasBenchmarkBase):
     default_flags['num_gpus'] = 1
     default_flags['train_epochs'] = 8
     default_flags['batch_size'] = 99000
-    default_flags['eval_batch_size'] = 160000
     default_flags['learning_rate'] = 0.00382059
     default_flags['beta1'] = 0.783529
     default_flags['beta2'] = 0.909003

official/recommendation/ncf_keras_main.py

@@ -66,16 +66,10 @@ class MetricLayer(tf.keras.layers.Layer):
     self.params = params
     self.metric = tf.keras.metrics.Mean(name=rconst.HR_METRIC_NAME)
 
-  def call(self, inputs, training=False):
+  def call(self, inputs):
     logits, dup_mask = inputs
-
-    if not training:
     in_top_k, metric_weights = metric_fn(logits, dup_mask, self.params)
-      metric = self.metric(in_top_k, sample_weight=metric_weights)
-    else:
-      metric = 0.0
-
-    self.add_metric(metric, name="ncf_metric", aggregation="mean")
+    self.add_metric(self.metric(in_top_k, sample_weight=metric_weights))
     return logits
 
 
@@ -255,7 +249,7 @@ def run_ncf(_):
     (train_input_dataset, eval_input_dataset,
      num_train_steps, num_eval_steps) = \
       (ncf_input_pipeline.create_ncf_input_data(
-          params, producer, input_meta_data, strategy))
+          params, producer, input_meta_data))
     steps_per_epoch = None if generate_input_online else num_train_steps
 
     with distribution_utils.get_strategy_scope(strategy):