-
DeepCTR/examples/run_dsin.py
Could you please clarify what sess_number in line 24?
Cheers
-
你好，请问下DIN，XDeepFM都会有multivalent 分类变量 embedding，这样的特征模型要怎么输入尼
-
在跑deepctr例子中
原有的一段代码是这样，
emb1 = xdl.embedding('emb1', batch['sparse0'], xdl.TruncatedNormal(stddev=0.001), 8, 1024, vtype='hash')
个人是想不经过hash，改成了下面的代码
emb1 = xdl.embedding('emb1', batch['sparse0']…
-
**Describe the question(问题描述)**
When I use ```pip install deepctr```, I found the arguments of DeepFM() is not consistent with those of source code. Which version should I use?
**Operating environ…
-
Hello,
First thank you for very useful lib !
I'm trying to predict CTR with Avazu dataset. I'm raising this issue because I have an unexpected error when I want to train my Convolutional CLick Pre…
-
使用local模式测试ps_mark_op, 在程序结束释放内存时出core 。core并非稳定复现，但是90%以上的情况都出，堆栈可能不同。另外在程序运行的中间也可能出问题，提示：corrupted size vs. prev_size。
分布式场景没有测过，不确定有没有类似问题。麻烦帮忙看看。
代码如下(使用的deepctr例子的样本数据)：
def train(option…
-
**Describe the question(问题描述)**
@shenweichen 非常感谢提供了CTR方面的模型实现！对于sparse变量和sequence变量，都能完全理解embedding的做法，但是对于dense变量，做embedding，不能完全理解merge_dense_input函数的意义，函数返回的embed_list，只是为了确保和sparse和sequence变量保持输…
-
2019-01-21 17:47:57 INFO XLearningContainer: Backtrace:
2019-01-21 17:47:57 INFO XLearningContainer: 0x0000000000846d5c
2019-01-21 17:47:57 INFO XLearningContainer: 0x0000000000847088
2019-01-2…
-
(gdb) bt
#0 std::_Head_base::_Head_base (__h=..., this=0x8) at /usr/include/c++/5/tuple:115
#1 std::_Tuple_impl::_Tuple_impl (__head=..., this=0x8) at /usr/include/c++/5/tuple:362
#2 std::tuple:…
-
**Describe the question(问题描述)**
It's difficult for me to understand doing embedding for a dense feature,Why we need do that?thx !
**Additional context**
Add any other context about the problem he…