Data Science on DA-Code (test)
34.8Data Wrangling ScoreDA-Agent
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| DA-AgentModel=Gemini-2.5-Pro, Framework=DA-Agent2025.09 | 34.8 | 57.2 | 31.1 | 50 | 34.2 | 32 | 37 | |
| DA-AgentModel=GPT-4, Framework=DA-Agent2025.09 | 30.4 | 48.4 | 24.6 | 45.4 | 27.8 | 23.4 | 30.5 | |
| DS-STARModel=Gemini-2.5-Pro, Framework=DS-STAR2025.09 | 30.4 | 57.3 | 34.8 | 48.9 | 35.2 | 37.1 | 38.5 | |
| DA-AgentModel=Claude-3-Opus, Framework=DA-Agent2025.09 | 29.3 | 46.8 | 20.7 | 44.7 | 23.8 | 19 | 27.6 | |
| AutoGenModel=Gemini-2.5-Pro, Framework=AutoGen2025.09 | 25.6 | 51.6 | 25.6 | 38.7 | 28.5 | 29.6 | 30.8 | |
| DA-AgentModel=DeepSeek-Coder-V2.5, Framework=DA-Agent2025.09 | 25.1 | 34.1 | 14.7 | 32.8 | 18.7 | 14.1 | 20.7 | |
| DA-AgentModel=Qwen-2.5-72B, Framework=DA-Agent2025.09 | 24.9 | 41.8 | 15.4 | 31.9 | 19.4 | 22.3 | 22.6 | |
| DA-AgentModel=Mixtral-8x22B, Framework=DA-Agent2025.09 | 14.8 | 31.6 | 10.2 | 17.6 | 16.8 | 8.6 | 15.4 | |
| ReActModel=Gemini-2.5-Pro, Framework=ReAct2025.09 | 14.7 | 31.2 | 22.2 | 32 | 17.9 | 25.9 | 22.5 |