search

chenchongthu / DeepCoNN

This is our implementation of DeepCoNN
151 stars 65 forks source link

Dropout rate #20

Closed sh0416 closed 4 years ago

sh0416 commented 4 years ago

Can I ask you why you use drop rate with 1.0??

I am new to tensorflow, and I found that it means you make all your feature goes to zero.

I don't understand why you choose this..

with tf.name_scope("dropout"):
            self.h_drop_u = tf.nn.dropout(self.h_pool_flat_u, 1.0)
            self.h_drop_i= tf.nn.dropout(self.h_pool_flat_i, 1.0)

in model/deepconn.py

Thanks,

sh0416 commented 4 years ago

It has different api for the tensorflow version, close issue..