tensorflow-DeepFM
tensorflow-DeepFM copied to clipboard
Tensorflow implementation of DeepFM for CTR prediction.
模型auc
看着模型的auc 似乎挺低的,只有0.64左右,请问这个符合预期吗? y label is: [0, 0, 0, 0, 0, 0, 0, 0, 0, 0] y pred is: [0.05798095 0.03533146 0.01641405 0.03143436 0.03862092 0.03267911 0.05392143 0.02281025 0.01579049 0.01842922] y label...
Hi, do you have a recommendation of tensor flow version? Further, if you tell us the version of CUDA and cuDNN, it is highly appreciated.
换了一批其他数据,然后dfm.predict 结果为nan,求大神指点
关于代码107行
deepFM.py 第107行为:self.y_deep = tf.nn.dropout(self.y_deep, self.dropout_keep_deep[0]),本人在实验中,删除该行,效果反而变好了很多。查阅资料后,发现dropout一般的对象是deep的中间节点,而该行是针对初始节点。 楼主可以自行实验,以加验证。 祝好~
From the equation (2) in the paper, the second order term should be a scalar. However, in the code, it is computed as a vector. I think the line 102...