def test_distribution_overflow(self, seq_conf):
target_rate = 0.9999999
seq_conf = synthetic.Config((1, 1), 5, 1, rate=target_rate)
seq = self.cls(seq_conf)
rate = analysis.event_rate(seq)
def test_0_rate(self, seq_conf):
target_rate = 0.0
seq_conf = synthetic.Config((3, 3), 4, 1, rate=target_rate)
seq = self.cls(seq_conf)
rate = analysis.event_rate(seq)
assert rate == target_rate
def test_rates(self, seq_conf):
target_rates = [0.4235971, 0.073911]
for target_rate in target_rates:
seq_conf = synthetic.Config((32, 32), 500, 1, rate=target_rate)
seq = self.cls(seq_conf)
rate = analysis.event_rate(seq)
assert target_rate == pytest.approx(rate, 0.001)
def test_computed_rate(self, seq_conf):
target_rate = 0.4235971
res = 30
frames = 500
seq_conf = synthetic.Config((res, res), frames, 1, rate=target_rate)
seq = self.cls(seq_conf)
rate = analysis.event_rate(seq)
assert seq.rate == rate
assert seq.rate != target_rate
def test_event_rate(): seq = [ np.array([[0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]], dtype='uint8'), np.array([[1, 1, 1, 1], [0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]], dtype='uint8'), np.array([[2, 2, 2, 2], [0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]], dtype='uint8'), np.array([[3, 3, 3, 3], [0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]], dtype='uint8') ] result = ev.event_rate(seq) assert result == 0.25
def analyze(video_path):
seq = GrayscaleVideoConverter(VideoSequence(video_path))
return event_rate(seq)