The test test_matching_log_probas sometimes fails when the assertion on line 86 fails. It failed 2/200 times for me. I think this is happening due to a rounding error in the log probabilities. As a result sometimes the value is rounded up to 0 causing the assertion to fail.
Re-running the test would reduce the failure rate to less than 1%. Do you guys think this makes sense? I will be happy to incorporate any suggestions that you guys may have.
The test
test_matching_log_probassometimes fails when the assertion on line 86 fails. It failed 2/200 times for me. I think this is happening due to a rounding error in the log probabilities. As a result sometimes the value is rounded up to 0 causing the assertion to fail.Re-running the test would reduce the failure rate to less than 1%. Do you guys think this makes sense? I will be happy to incorporate any suggestions that you guys may have.
Thanks!