From aa95aa21afc585f643ae9cd6c732dcb1ba2d3a9b Mon Sep 17 00:00:00 2001 From: Rita Aleksziev Date: Tue, 12 Nov 2024 17:40:42 +0100 Subject: [PATCH] downloading example repo for eval --- evals/{eval_on_swe_bench.py => deepeval_on_swe_bench.py} | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) rename evals/{eval_on_swe_bench.py => deepeval_on_swe_bench.py} (97%) diff --git a/evals/eval_on_swe_bench.py b/evals/deepeval_on_swe_bench.py similarity index 97% rename from evals/eval_on_swe_bench.py rename to evals/deepeval_on_swe_bench.py index b5b3f14b6..b191a08d4 100644 --- a/evals/eval_on_swe_bench.py +++ b/evals/deepeval_on_swe_bench.py @@ -160,8 +160,8 @@ if __name__ == "__main__": # run_cognify_base_rag_and_search() # # Data preprocessing before setting the dataset test cases swe_dataset = load_swebench_dataset('princeton-nlp/SWE-bench_bm25_13K', split='test') - test_dataset = convert_swe_to_deepeval_testcases(dataset.test_cases) + test_dataset = convert_swe_to_deepeval_testcases(swe_dataset) from deepeval.metrics import HallucinationMetric metric = HallucinationMetric() - evalresult = dataset.evaluate([metric]) + evalresult = test_dataset.evaluate([metric]) pass \ No newline at end of file