Hello, I found a performance issue in the definition of cudnn_gru, MRC/BiDAF/layers.py, tf.zeros([1, batch_size, num_units]) will be created repeatedly during program execution, resulting in reduced efficiency. I think it should be created before the loop.
Hello, I found a performance issue in the definition of cudnn_gru, MRC/BiDAF/layers.py,
tf.zeros([1, batch_size, num_units])will be created repeatedly during program execution, resulting in reduced efficiency. I think it should be created before the loop.The same issue exist in :
tf.zerosin line 97;tf.zerosin line 149 and 150;tf.zerosin line 131 and 132;tf.zerosin line 187 and 188;tf.zerosin line 133 and 134;tf.zerosin line 186 and 187;tf.zerosin line 235 and 236;Looking forward to your reply. Btw, I am very glad to create a PR to fix it if you are too busy.