def test_JaccardMetric():
# set up synthentic data for testing
targets = np.array([
[True, True, False], [True, False, True], [False, True, False],
[True, True, True], [False, False, True]
])
preds = np.array([
[1, 0, 0], [0, 1, 1], [0, 1, 0],
[1, 1, 1], [0, 0, 1]
])
# create instance of jaccard metric object
jm = JaccardMetric()
# first test all the data at once
jm.update_state(targets, preds)
jaccard_big_batch = jm.result().numpy()
# reset object and test multiple matches
jm.reset_states()
# Do two mini-batches
jm.update_state(targets[:3], preds[:3])
jm.update_state(targets[3:], preds[3:])
jaccard_multiple_batches = jm.result().numpy()
# check to make sure answers same for both big batch and multiple batches
assert np.isclose(jaccard_big_batch, jaccard_multiple_batches)
def _setup_metrics(self):
self.metric_functions = {} # needed to shadow class variable
self.metric_functions[LOSS] = self.eval_loss_function
self.metric_functions[JACCARD] = JaccardMetric()