search

yuanzhoulvpi2017 / quick_sentence_transformers

sentence-transformers to onnx 让sbert模型推理效率更快
MIT License
156 stars 30 forks source link

cpu下加速 #6

Open ZTurboX opened 7 months ago

ZTurboX commented 7 months ago

测试下来，cpu下没有加速，反而比原始sentence-transformer要慢，这是什么原因呢

yuanzhoulvpi2017 commented 7 months ago

cpu没办法优化，肯定是要在gpu上优化的