How it works?

[Zhengxuru]

bilm-tf/bilm/model.py

Line 547 in 7cffee2

tf.Variable(

lstm_cell.state_size is an Integer or TensorShape, what does "for dim in lstm_cell.state_size" mean? And I don't know how "state_update_op = tf.assign(init_states[i], new_state)" works, since you assign init_states using "tf.zeros([self._max_batch_size, dim]" in L548.