30 lines
941 B
JSON
30 lines
941 B
JSON
|
{
|
||
|
|
"command": "agbenchmark start --test TestDebugSimpleTypoWithGuidance",
|
||
|
|
"completion_time": "2023-07-18-07:46",
|
||
|
|
"metrics": {
|
||
|
|
"run_time": "86.86 seconds",
|
||
|
|
"highest_difficulty": "novice: 3"
|
||
|
|
},
|
||
|
|
"tests": {
|
||
|
|
"TestDebugSimpleTypoWithGuidance": {
|
||
|
|
"data_path": "agbenchmark/challenges/code/d1_debug",
|
||
|
|
"is_regression": false,
|
||
|
|
"task": "1- Run test.py.\n2- Read code.py.\n3- Modify code.py.\nRepeat step 1, 2 and 3 until test.py runs without errors.\n",
|
||
|
|
"answer": "[0, 1] [2, 5] [0, 3]",
|
||
|
|
"description": "Tests ability for the agent to debug python code with a simple typo in it.",
|
||
|
|
"metrics": {
|
||
|
|
"difficulty": "novice",
|
||
|
|
"success": true,
|
||
|
|
"success_%": 100.0,
|
||
|
|
"run_time": "86.579 seconds"
|
||
|
|
},
|
||
|
|
"reached_cutoff": true
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"config": {
|
||
|
|
"workspace": "${os.path.join(Path.home(), 'miniagi')}"
|
||
|
|
},
|
||
|
|
"additional": {
|
||
|
|
"model": "gpt-3.5-turbo"
|
||
|
|
}
|
||
|
|
}
|