Internal change

PiperOrigin-RevId: 263158478

official/recommendation/ncf_keras_benchmark.py

@@ -302,6 +302,19 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.force_v2_in_keras_compile = False
     self._run_and_report_benchmark_mlperf_like()
 
+  def benchmark_xla_8_gpu_mlperf_like(self):
+    """8 GPU using keras fit/compile with XLA."""
+    self._setup()
+    FLAGS.num_gpus = 8
+    FLAGS.enable_xla = True
+    FLAGS.train_epochs = 17
+    FLAGS.batch_size = 1048576
+    FLAGS.learning_rate = 0.0045
+    FLAGS.beta1 = 0.25
+    FLAGS.beta2 = 0.5
+    FLAGS.epsilon = 1e-8
+    self._run_and_report_benchmark_mlperf_like()
+
   def benchmark_8_gpu_ctl_mlperf_like(self):
     """8 GPU using CTL."""
     self._setup()
@@ -315,6 +328,20 @@ class NCFKerasAccuracy(NCFKerasBenchmarkBase):
     FLAGS.epsilon = 1e-8
     self._run_and_report_benchmark_mlperf_like()
 
+  def benchmark_xla_8_gpu_ctl_mlperf_like(self):
+    """8 GPU using CTL with XLA."""
+    self._setup()
+    FLAGS.keras_use_ctl = True
+    FLAGS.enable_xla = True
+    FLAGS.num_gpus = 8
+    FLAGS.train_epochs = 17
+    FLAGS.batch_size = 1048576
+    FLAGS.learning_rate = 0.0045
+    FLAGS.beta1 = 0.25
+    FLAGS.beta2 = 0.5
+    FLAGS.epsilon = 1e-8
+    self._run_and_report_benchmark_mlperf_like()
+
 
 class NCFKerasSynth(NCFKerasBenchmarkBase):
   """Benchmark NCF model using synthetic data."""