def test_censoring_funs_no_time_discrete(self, expected_tte_d,
expected_is_censored_d, events_d):
times_to_event = padded_events_to_tte(events_d, discrete_time=True)
not_censored = padded_events_to_not_censored(events_d,
discrete_time=True)
assert (
expected_tte_d == times_to_event).all(), ' time_to_event failed'
assert (expected_is_censored_d !=
not_censored).all(), 'not_censored failed'
def test_censoring_funs_no_time():
# TODO proper unit testing
# print 'TTE & CENSORING'
# print 'padded discrete'
expected_tte = expected_tte_d
expected_is_censored = expected_is_censored_d
times_to_event = padded_events_to_tte(events_d, discrete_time=True)
not_censored = padded_events_to_not_censored(events_d, discrete_time=True)
assert (expected_tte == times_to_event).all(), ' time_to_event failed'
assert (expected_is_censored != not_censored).all(), 'not_censored failed'
# print 'padded continuous'
expected_tte = expected_tte_c
expected_is_censored = expected_is_censored_c
times_to_event = padded_events_to_tte(events_c, discrete_time=False)
not_censored = padded_events_to_not_censored(events_c, discrete_time=False)
assert (expected_tte == times_to_event).all(), ' time_to_event failed'
assert (expected_is_censored != not_censored).all(), 'not_censored failed'
def test_censoring_funs_no_time_continuous(self, expected_tte_c,
expected_is_censored_c,
events_c):
times_to_event = padded_events_to_tte(events_c, discrete_time=False)
not_censored = padded_events_to_not_censored(events_c,
discrete_time=False)
assert (
expected_tte_c == times_to_event).all(), ' time_to_event failed'
assert (expected_is_censored_c !=
not_censored).all(), 'not_censored failed'
def prep_tensors(x,events):
# 0. calculate time to event and censoring indicators.
y = np.ones([events.shape[0],events.shape[1],2])
y[:,:,0] = tr.padded_events_to_tte(np.squeeze(events),discrete_time=True)
y[:,:,1] = tr.padded_events_to_not_censored(np.squeeze(events),discrete_time=True)
# 1. Disalign features and targets otherwise truth is leaked.
# 2. drop first timestep (that we now dont have features for)
# 3. nan-mask the last timestep of features. (that we now don't have targets for)
events = events[:,1:,]
y = y[:,1:]
x = np.roll(x, shift=1, axis=1)[:,1:,]
x = x + 0*np.expand_dims(events,-1)
return x,y,events