[Bug]: Bug of Evaluation for Trae Agent.
What happened?
This bug was occured while I am evaluating Swebbench-Verified problems with Trae Agent. I insalled all moudles and cloned swebench github, just followed what was written here.https://github.com/bytedance/trae-agent/blob/main/evaluation/README.md
I got this error while I am running this command: python run_evaluation.py --instance_ids django__django-11138
Traceback (most recent call last): File "/root/Trae/trae-agent/evaluation/run_evaluation.py", line 20, in <module> from .utils import BENCHMARK_CONFIG, docker_exec ImportError: attempted relative import with no known parent package
So I needed to go back fomer directory and had to run this command: python -m evaluation.run_evaluation --instance_ids django__django-11138
It worked but got some error like: docker and pexpect module not found.
So I update the code on my end and sent PR for this issue. https://github.com/bytedance/trae-agent/pull/370
But still got another issue.
Checking image status: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 500/500 [00:00<00:00, 909.69it/s] Total number of images: 1 Number of images to download: 0 Found built trae-agent and uv artifacts. Skipping building. Running specified instances: ['django__django-11138'] Running instances: 100%|████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [05:25<00:00, 325.32s/it] Traceback (most recent call last): File "<frozen runpy>", line 198, in _run_module_as_main File "<frozen runpy>", line 88, in _run_code File "/root/Trae/trae-agent/evaluation/run_evaluation.py", line 475, in <module> main() File "/root/Trae/trae-agent/evaluation/run_evaluation.py", line 471, in main evaluation.run_eval() File "/root/Trae/trae-agent/evaluation/run_evaluation.py", line 343, in run_eval process = subprocess.run(cmd, capture_output=True, cwd=benchmark_harness_path.as_posix()) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.12/subprocess.py", line 548, in run with Popen(*popenargs, **kwargs) as process: ^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/lib/python3.12/subprocess.py", line 1026, in __init__ self._execute_child(args, executable, preexec_fn, close_fds, File "/usr/lib/python3.12/subprocess.py", line 1955, in _execute_child raise child_exception_type(errno_num, err_msg, err_filename) FileNotFoundError: [Errno 2] No such file or directory: 'swebench_venv/bin/python'
Not sure why this happened.
What did you expect to happen?
It would be super great if you check this issue and solve it asap. Thanks.
Traceback
No response
What is your system, Python, dependency version?
No response
Additional information that you believe is relevant to this bug
No response
Source: bytedance/trae-agent