Closed fsaad closed 8 months ago
evaluate_test.py
test_evaluat_mini.py
Failling tests appear to arise from disagreement between predictions on the Github Actions server and predictions on local machine, which are stored in these files:
evaluate_test.py
in favor of a fastertest_evaluat_mini.py
, which takes a somewha reasonable 1min.