Did you use clinc150 data_imbalanced vs data_full?

Hi,

I experimented with verifying your approach however I didn't receive such low F1 numbers for few shot. In looking at data_full, banking has the same amount of training samples as all other domains/intents. Did you in fact use data imbalanced?

将报告中的 CLINC150 实验与 repository 的 data_full 和 data_imbalanced 数据集进行比较，重点关注 banking 领域和训练样本数量。首先确定实验使用了哪个数据集，然后验证 few-shot F1 结果。完成的标准是记录清楚该差异是否源于数据集选择，或源于其他可复现的设置差异。

由索引模型根据 Issue 内容生成。