OpenHands/evaluation/swe_bench

SWE-Bench Evaluation with OpenDevin SWE-Bench Docker Image

This folder contains evaluation harness we built on top of the original SWE-Bench benchmark (paper). We create a fork of SWE-Bench mostly build on top of the original repo and containerized it for easy evaluation.

OpenDevin SWE-Bench Docker Image

In OpenDevin-SWE-Bench fork (mostly from original repo with some fixes), we try to pre-build the testbed (i.e., code of the repository we want the agent to edit) AND the conda environment, so that in evaluation (inference) time, we can directly leverage existing environments for effecienct evaluation.

We pack everything you need for SWE-Bench evaluation into one, gigantic, docker image. To use it:

docker pull ghcr.io/opendevin/eval-swe-bench:full-v1.0

The Docker image contains several important directories:

• /swe_util/OD-SWE-bench: root directory for the OD-SWE-bench repository
• /swe_util/eval_data: director to eval data
  • /swe_util/eval_data/eval_logs/: evaluation logs
  • /swe_util/eval_data/eval_temp/: temporary folder for the evaluation process
  • /swe_util/eval_data/instances/: swe-bench raw instances
  • /swe_util/eval_data/outputs/: model or agent outputs
  • /swe_util/eval_data/testbed_logs/: logs for testbed building
  • /swe_util/eval_data/testbeds/: directory for all testbeds
• /swe_util/miniforge3/: directory for miniforge3

To reproduce how we pack the image, check this doc.

NOTE: We only support SWE-Bench lite for now. But modifying our existing scripts for full SWE-Bench should be quite straight forward.

Configure OpenDevin and your LLM

Create a config.toml file if it does not exist at the root of the workspace.

Add the following configurations:

[core]
max_iterations = 100
cache_dir = "/tmp/cache"
sandbox_container_image = "ghcr.io/opendevin/sandbox:latest"
sandbox_type = "ssh"
use_host_network = true
ssh_hostname = "localhost"
sandbox_timeout = 120
# SWEBench eval specific
run_as_devin = false

# TODO: Change these to the model you want to evaluate
[eval_gpt4_1106_preview]
model = "gpt-4-1106-preview"
api_key = "XXX"
temperature = 0.0

[eval_some_openai_compatible_model]
model = "openai/MODEL_NAME"
base_url = "https://OPENAI_COMPATIBLE_URL/v1"
api_key = "XXX"
temperature = 0.0

Test if your environment works

python3 evaluation/swe_bench/swe_env_box.py

If you get to the interactive shell successfully, it means success!

Run Inference on SWE-Bench Instances

./evaluation/swe_bench/scripts/run_infer.sh eval_gpt4_1106_preview

You can replace eval_gpt4_1106_preview with any model you setted up in config.toml.

Evaluate Generated Patches

After running the inference described in the previous section, you will obtain a output.jsonl (by default it will save to evaluation/evaluation_outputs). Then you can run this one line script to evaluate generated patches, and produce a fine-grained report:

If you want to evaluate existing results, you should first run this to clone existing outputs

git clone https://huggingface.co/spaces/OpenDevin/evaluation evaluation/evaluation_outputs

Then you can run the following:

# ./evaluation/swe_bench/scripts/eval_infer.sh $YOUR_OUTPUT_JSONL
# For example:
./evaluation/swe_bench/scripts/eval_infer.sh evaluation/evaluation_outputs/outputs/swe_bench/CodeActAgent/gpt-4-1106-preview_maxiter_50_N_v1.0/output.jsonl

The final results will be saved to evaluation/evaluation_outputs/outputs/swe_bench/CodeActAgent/gpt-4-1106-preview_maxiter_50_N_v1.0/output.merged.jsonl.

It will contains an additional field fine_grained_report (see example below) compared to the output.jsonl from the previous inference stage.

"fine_grained_report": {
  "gold_tests": {
    "FAIL_TO_PASS": "[\"tests/test_ext_viewcode.py::test_viewcode_epub_default\"]",
    "PASS_TO_PASS": "[\"tests/test_ext_viewcode.py::test_viewcode_epub_enabled\", \"tests/test_ext_viewcode.py::test_linkcode\", \"tests/test_ext_viewcode.py::test_local_source_files\"]"
  },
  "generated": true,
  "with_logs": true,
  "applied": true,
  "test_errored": false,
  "test_timeout": false,
  "resolved": true,
  "log_parse": {
    "tests/test_ext_viewcode.py::test_viewcode_epub_default": "PASSED",
    "tests/test_ext_viewcode.py::test_viewcode_epub_enabled": "PASSED",
    "tests/test_ext_viewcode.py::test_linkcode": "PASSED",
    "tests/test_ext_viewcode.py::test_local_source_files": "PASSED",
    "tests/test_ext_viewcode.py::test_viewcode": "FAILED"
  },
  "eval_report": {
    "FAIL_TO_PASS": {
      "success": [
        "tests/test_ext_viewcode.py::test_viewcode_epub_default"
      ],
      "failure": []
    },
    "PASS_TO_PASS": {
      "success": [
        "tests/test_ext_viewcode.py::test_viewcode_epub_enabled",
        "tests/test_ext_viewcode.py::test_linkcode",
        "tests/test_ext_viewcode.py::test_local_source_files"
      ],
      "failure": []
    },
    "FAIL_TO_FAIL": {
      "success": [],
      "failure": []
    },
    "PASS_TO_FAIL": {
      "success": [],
      "failure": []
    }
  }
}

Please refer to EVAL_PATCH.md if you want to learn more about how to evaluate patches that are already generated (e.g., not by OpenDevin).

Submit your evaluation results

You can start your own fork of our huggingface evaluation outputs and submit a PR of your evaluation results following the guide here.