You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
It's a little trick. We found empirically that the performance is better via splitting direct connections (to output unit) and hidden layers to next stage.
hparams.logger.info("split connect")
if idx != len(hparams.cross_layer_sizes) - 1:
next_hidden, direct_connect = tf.split(curr_out, 2 * [int(layer_size / 2)], 1)
final_len += int(layer_size / 2)
else:
direct_connect = curr_out
next_hidden = 0
final_len += layer_size
field_nums.append(int(layer_size / 2))
I want to know why does "next_hidden, direct_connect = tf.split(curr_out, 2 * [int(layer_size / 2)], 1)" this action
The text was updated successfully, but these errors were encountered: