Type Prediction On Manytypes4Typescript
评估指标
Average Accuracy
Average F1
Average Precision
Average Recall
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Average Accuracy | Average F1 | Average Precision | Average Recall |
|---|---|---|---|---|
| multilingual-training-for-software | 61.29 | 58.86 | 58.81 | 58.91 |
| 模型 2 | 59.81 | 56.71 | 56.57 | 56.85 |
| roberta-a-robustly-optimized-bert-pretraining | 59.84 | 57.54 | 57.45 | 57.62 |
| bert-pre-training-of-deep-bidirectional | 57.52 | 54.10 | 54.18 | 54.02 |
| manytypes4typescript-a-comprehensive | 63.42 | - | - | - |
| graphcodebert-pre-training-code | 62.51 | 60.57 | 60.06 | 61.08 |
| learning-type-inference-for-enhanced-dataflow | 71.27 | - | - | - |
| multilingual-training-for-software | 61.00 | 58.63 | 58.36 | 58.91 |
| codebert-a-pre-trained-model-for-programming | 61.72 | 59.57 | 59.34 | 59.80 |