Save and load state of tf.random

I have a Tensorflow model that utilises dropout, which is controlled by the state of tf.random. At the start of my code, I set the global seed using tf.set_seed(DEFAULT_SEED). As training progresses, I want to be able to save the state of tf.random, so that if I want to stop and then resume training, I can load that state and training can continue from where it left off. For example, np.random has methods set_state() and get_state(), however I can't find an equivalent for tf.random. Am I missing something or is this not doable?

Numpy example:

np.random.set_seed(1234)

for i in range(10):
    print(np.random.randn())

x = np.random.get_state()
np.random.set_state(x)

and the state is the same as where we left it.