提交 eb49ae73 编写于 作者: H Hongkun Yu 提交者: A. Unique TensorFlower

Add NHNet cloud tpu accuracy benchmark

PiperOrigin-RevId: 312596462
上级 241c985a
......@@ -31,7 +31,7 @@ from official.utils.flags import core as flags_core
MIN_LOSS = 0.35
MAX_LOSS = 0.45
NHNET_DATA = 'nhnet_training_data_files'
NHNET_DATA = 'gs://tf-perfzero-data/nhnet/v1/train.tfrecord*'
FLAGS = flags.FLAGS
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册