Skip to content

P
Paddle

机器未来 / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
提交 56fbed9a 编写于 作者: D dongzhihong
浏览文件
操作

"fix dense vector shape in mq2007"

上级 45ce1649
显示空白变更内容
内联 并排
Showing with 3 addition and 3 deletion
python/paddle/v2/dataset/mq2007.py
浏览文件 @ 56fbed9a
...@@ -212,19 +212,19 @@ def gen_pair(querylist, partial_order="full"): ...@@ -212,19 +212,19 @@ def gen_pair(querylist, partial_order="full"):
for j in range(i + 1, len(querylist)): for j in range(i + 1, len(querylist)):
query_right = querylist[j] query_right = querylist[j]
if query_left.relevance_score > query_right.relevance_score: if query_left.relevance_score > query_right.relevance_score:
labels.append(1) labels.append([1])
docpairs.append([ docpairs.append([
np.array(query_left.feature_vector), np.array(query_left.feature_vector),
np.array(query_right.feature_vector) np.array(query_right.feature_vector)
]) ])
elif query_left.relevance_score < query_right.relevance_score: elif query_left.relevance_score < query_right.relevance_score:
labels.append(1) labels.append([1])
docpairs.append([ docpairs.append([
np.array(query_right.feature_vector), np.array(query_right.feature_vector),
np.array(query_left.feature_vector) np.array(query_left.feature_vector)
]) ])
for label, pair in zip(labels, docpairs): for label, pair in zip(labels, docpairs):
yield label, pair[0], pair[1] yield np.array(label), pair[0], pair[1]
def gen_list(querylist): def gen_list(querylist):
......
Markdown is supported
0% 或 .
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 或