lapis-hong / wide-resdnn Goto Github PK

View Code? Open in Web Editor NEW

82.0 82.0 27.0 3.88 MB

Wide and Deep Learning(Wide&ResDNN) for Kaggle Criteo Dataset in tensorflow

Shell 4.76% Python 95.24%

ctr-prediction deep-learning tensorflow wide-and-deep

wide-resdnn's People

Contributors

Stargazers

Watchers

wide-resdnn's Issues

wide_dim

In your code, 'wide_dim += hash_bucket_size'.

If all 39 features are used for wide part, then the number of cross feature should be very large, maybe 1000**39. So the above code should be 'wide_dim *= hash_bucket_size'

In the tf tutorial of wide & deep, only some features are selected for wide part. I wonder which to choose in criteo dataset? thx~

when run train.py have a error, Expect 40 fields but have 39 in record 0,how to solve it?

INFO:tensorflow:Starting evaluation ……
tensorflow.python.framework.errors_impl.InvalidArgumentError: Expect 40 fields but have 39 in record 0

The graph couldn't be sorted in topological order。

Hello,there is an error when I run your code: topological sort failed with message: The graph couldn't be sorted in topological order.
I do not know why this error occured. Can you explain this problem for me? Thank you very much.

Recommend Projects