First pass at adding hypothesis generated test data. Replaced np.random calls in property-based tests with the corresponding hypothesis calls. Have to use deadline=None because of numba (first run is slow compared to the subsequent runs and that causes hypothesis to fail the test for being "flaky").
First pass at adding hypothesis generated test data. Replaced np.random calls in property-based tests with the corresponding hypothesis calls. Have to use
deadline=None
because of numba (first run is slow compared to the subsequent runs and that causes hypothesis to fail the test for being "flaky").