未验证 提交 49ad570e 编写于 作者: T Tao Luo 提交者: GitHub

Merge pull request #12667 from luotao1/wmt14

fix wmt14 download error
......@@ -36,8 +36,7 @@ URL_DEV_TEST = ('http://www-lium.univ-lemans.fr/~schwenk/'
MD5_DEV_TEST = '7d7897317ddd8ba0ae5c5fa7248d3ff5'
# this is a small set of data for test. The original data is too large and
# will be add later.
URL_TRAIN = ('http://paddlepaddle.cdn.bcebos.com/demo/'
'wmt_shrinked_data/wmt14.tgz')
URL_TRAIN = ('http://paddlemodels.bj.bcebos.com/wmt/wmt14.tgz')
MD5_TRAIN = '0791583d57d5beb693b9414c5b36798c'
# BLEU of this trained model is 26.92
URL_MODEL = 'http://paddlemodels.bj.bcebos.com/wmt%2Fwmt14.tgz'
......
......@@ -15,7 +15,7 @@
WMT14 dataset.
The original WMT14 dataset is too large and a small set of data for set is
provided. This module will download dataset from
http://paddlepaddle.cdn.bcebos.com/demo/wmt_shrinked_data/wmt14.tgz and
http://paddlemodels.bj.bcebos.com/wmt/wmt14.tgz and
parse training set and test set into paddle reader creators.
"""
......@@ -37,8 +37,7 @@ URL_DEV_TEST = ('http://www-lium.univ-lemans.fr/~schwenk/'
MD5_DEV_TEST = '7d7897317ddd8ba0ae5c5fa7248d3ff5'
# this is a small set of data for test. The original data is too large and
# will be add later.
URL_TRAIN = ('http://paddlepaddle.cdn.bcebos.com/demo/'
'wmt_shrinked_data/wmt14.tgz')
URL_TRAIN = ('http://paddlemodels.bj.bcebos.com/wmt/wmt14.tgz')
MD5_TRAIN = '0791583d57d5beb693b9414c5b36798c'
# BLEU of this trained model is 26.92
URL_MODEL = 'http://paddlemodels.bj.bcebos.com/wmt%2Fwmt14.tgz'
......
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