THUDM / P-tuning-v2

An optimized deep prompt tuning strategy comparable to fine-tuning across scales and tasks
Apache License 2.0
1.96k stars 198 forks source link

多任务学习的训练代码在哪呢没找到呀~ #31

Closed Torres9999 closed 2 years ago

Xiao9905 commented 2 years ago

@Torres9999 Hi,

The multitask learning code is under construction. You may refer to #22 for details if you wanna implement it yourself.