跳到主要内容

evaluation.benchmarks.mint.utils

check_correctness​

def check_correctness(solution_code: str,
                      test_code: str,
                      timeout: float = 10,
                      completion_id: int | None = None) -> dict

Evaluates the functional correctness of a completion by running the test

suite provided in the problem.

Arguments:

  • completion_id: an optional completion ID so we can match the results later even if execution finishes asynchronously.

WriteOnlyStringIO Objects​

class WriteOnlyStringIO(io.StringIO)

StringIO that throws an exception when it's read from

readable​

def readable(*args, **kwargs)

Returns True if the IO object can be read.

reliability_guard​

def reliability_guard(maximum_memory_bytes: int | None = None)

This disables various destructive functions and prevents the generated code from interfering with the test (e.g. fork bomb, killing other processes, removing filesystem files, etc.)

Warnings:

This function is NOT a security sandbox. Untrusted code, including, model- generated code, should not be blindly executed outside of one. See the Codex paper for more information about OpenAI's code sandbox, and proceed with caution.