Method execute_code

src/evaluate.py:371–375

(self, code, io, testcases, check = True, fast_check = False, total_timeout = 10)

    def execute_code(self, code, io, testcases, check = True, fast_check = False, total_timeout = 10):
        if len(testcases) == 0:
            raise ValueError("No testcase to be executed.")
        stat, testcases = untrusted_check(io, code, testcases, 0, [30 for t in testcases], check = check, fast_check = True, total_timeout = total_timeout)
        return stat, testcases


    # unbiased estimator from https://github.com/openai/human-eval