issues
search
InsaneLife
/
dssm
DSSM and Multi-View DSSM
658
stars
230
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
测试准确度低
#37
jiakechong1991
opened
2 years ago
0
对bert输出的所有token做平均池化
#36
cxx-cindy
opened
2 years ago
0
损失函数的定义只涉及到了正样本？
#35
ninesky110
closed
3 years ago
2
dssm_rnn.py gpu利用率低
#34
lonelydancer
opened
3 years ago
1
train.py里的SiamenseBert系列是不是就没有负例了，跟dssm paper里的任务不一致了。
#33
lonelydancer
opened
3 years ago
1
🚨 Potential Deserialization of Untrusted Data
#32
huntr-helper
opened
3 years ago
1
mac或者linux可以训练吗
#31
travel-go
opened
3 years ago
1
[SEP]是不是需要加一下？
#30
hrwleo
closed
3 years ago
1
如何预测
#29
InsaneLife
closed
3 years ago
0
代码有误？
#28
sars715
closed
3 years ago
1
训练loss正常下降，验证loss下降缓慢
#27
mokii
opened
4 years ago
6
怎么做预测呀？
#26
JNero
closed
3 years ago
2
i want to predict a query pair like "今天天气怎么样" “今天温度怎么样” ，what to do?
#25
yangyan1227
closed
3 years ago
5
训练环境
#24
phoenixzxj
opened
5 years ago
0
dssm loss计算为什么是reduce_sum
#23
liamlmy
closed
4 years ago
0
训练dssm_rnn过程中，训练loss下降，为什么测试loss不下降反而增大
#22
mokii
opened
5 years ago
9
ModuleNotFoundError: No module named 'multi_view_data_input'
#21
ruirong92
opened
5 years ago
4
为何用数据里的label？label表示“是否点击”，不是相似性
#20
guotong1988
closed
5 years ago
2
效果怎么样？
#19
guotong1988
opened
5 years ago
0
The link of dataset cannot be retrieved，can you send me a new version ?
#18
xugege0523
opened
5 years ago
1
Why my train loss equal to nan?
#17
bingoohe
opened
5 years ago
2
pull batch使用的是验证级的数据
#16
MC-Zealot
closed
5 years ago
1
Readme中loss指标训练数据和测试数据是一样的吗？
#15
MC-Zealot
closed
5 years ago
1
请问您新加的loss curves所用的数据量是多少呢
#14
MC-Zealot
closed
5 years ago
1
ValueError: Cannot feed value of shape (0,) for Tensor 'input/query_batch:0', which has shape '(?, ?)'
#13
MC-Zealot
closed
5 years ago
1
data_input能发一份吗？
#12
sarary
closed
5 years ago
4
能说一下输入的query_in，positive_in，negative_in的shape吗
#11
haohuangch
closed
5 years ago
1
Data fromat
#10
FourierHe
closed
5 years ago
2
datasets format problem
#9
tzSmilence
closed
5 years ago
3
share data_input.py ？
#8
ablinke
closed
5 years ago
1
data_input
#7
zhaohongjie
closed
5 years ago
3
why my train loss ,after 4or5 epoch ,softmax value equal nan。
#6
chongqingwei
closed
5 years ago
6
为什么训练出来一个epoch，loss就位0
#5
chongqingwei
closed
5 years ago
4
方便写个输入数据的列子吗？
#4
yueyongjiao
closed
5 years ago
1
数据格式
#3
zhongyunuestc
closed
5 years ago
2
what the return of data_input.get_search_data
#2
SeekPoint
closed
5 years ago
2
more details on dataset format?
#1
SeekPoint
closed
5 years ago
1