Code Generation on DS-1000 1.0 (test)
67.2MatplotlibCodeQwen1.5-7B-Chat
Evaluation results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| CodeQwen1.5-7B-ChatTemperature=0.2, top-p=0.95, Samples=402024.10 | 67.2 | 51.7 | 34.4 | 52.8 | 38.9 | 47.9 | 46 | — | 47.1 | |
| CodeQwen1.5-7B-OctoPackTemperature=0.2, top-p=0.95, Samples=402024.10 | 61.8 | 42.6 | 26.7 | 31.4 | 32.7 | 36.6 | 37.7 | — | 38.2 | |
| StarCoder2-15BTemperature=0.2, top-p=0.95, Samples=402024.10 | 61.4 | 45.8 | 26.2 | 22.5 | 36 | 40.5 | 38.1 | — | 38.9 | |
| DeepSeek-Coder-6.7B-InstructTemperature=0.2, top-p=0.95, Samples=402024.10 | 58.4 | 51.1 | 34 | 50.6 | 34.2 | 45.8 | 45.9 | — | 44.6 | |
| SelfCodeAlign-CQ-7BTemperature=0.2, top-p=0.95, Samples=402024.10 | 57.2 | 42.6 | 28.2 | 33.3 | 35.6 | 42.8 | 38.3 | — | 39.1 | |
| CodeQwen1.5-7B-BaseTemperature=0.2, top-p=0.95, Samples=402024.10 | 56.7 | 35.9 | 21.6 | 23.8 | 28.2 | 30.9 | 28.8 | — | 32.4 | |
| CodeGemma-7B-ITTemperature=0.2, top-p=0.95, Samples=402024.10 | 54.7 | 34.4 | 21.9 | 34.5 | 21.8 | 22.6 | 25.1 | — | 30.8 | |
| OctoCoder-16BTemperature=0.2, top-p=0.95, Samples=402024.10 | 53.8 | 34 | 13.1 | 25.9 | 22.8 | 30 | 22.4 | — | 28.3 | |
| Llama-3-8B-InstructTemperature=0.2, top-p=0.95, Samples=402024.10 | 51.9 | 33.1 | 21.5 | 37.2 | 25.2 | 23.8 | 34.4 | — | 31.1 | |
| StarCoderFormat=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 51.7 | 29.7 | 11.4 | 21.4 | 20.2 | 29.5 | 24.5 | 26 | — | |
| StarCoderFormat=Insertion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 51.7 | 30.8 | 10.3 | 21 | 20.2 | 27.4 | 20 | 25.4 | — | |
| StarCoderBaseFormat=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 47 | 27.1 | 10.1 | 19.5 | 21.7 | 27 | 20.5 | 23.8 | — | |
| StarCoderBaseFormat=Insertion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 47 | 26.3 | 10.9 | 16.6 | 20.2 | 30.2 | 22.3 | 24 | — | |
| code-cushman-001Format=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 40.7 | 21.8 | 7.9 | 12.4 | 11.3 | 18 | 12.2 | 18.1 | — | |
| CodeGen-16B-MonoFormat=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 31.7 | 10.9 | 3.4 | 7 | 9 | 10.8 | 15.2 | 11.7 | — | |
| InCoder-6BFormat=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 28.3 | 4.4 | 3.1 | 4.4 | 2.8 | 2.8 | 3.8 | 7.4 | — | |
| InCoder-6BFormat=Insertion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 28.3 | 4.6 | 2.9 | 4.4 | 2.8 | 3.1 | 7.8 | 7.5 | — | |
| SantaCoder-1BFormat=Completion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 21.6 | 4.6 | 0.9 | 2.6 | 2.4 | 4.8 | 3.1 | 5.7 | — | |
| SantaCoder-1BFormat=Insertion, temperature=0.2, top_p=0.5, max_length=10242023.05 | 21.6 | 13.8 | 2 | 3.8 | 5.7 | 6.9 | 14.8 | 9.3 | — |