Feature Extraction
sentence-transformers
PyTorch
Chinese
English
bert
sentence-similarity
mteb
RAG
Eval Results (legacy)
text-embeddings-inference
Instructions to use DMetaSoul/Dmeta-embedding-zh with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use DMetaSoul/Dmeta-embedding-zh with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("DMetaSoul/Dmeta-embedding-zh") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Inference
- Notebooks
- Google Colab
- Kaggle
请问该模型目前是否只有STS任务的功能,不具备像BGE那样Instruct搜索的方式?
#3
by AiTutu - opened
比如在做RAG任务时,直接使用问句和文档片段的编码就可以搜索,还是要针对问句添加Instruction再编码?
1)模型离线优化时,不局限于 STS 任务,包括召回、精排、聚类等经典 NLP 语义理解任务均有涉及,所以在多场景、多任务上泛化能力极佳
2)进行RAG时,不需要考虑问句和文档片段的非对称问题,也即直接该模型同时编码问句和文档就行,不需要针对问句添加提示词了,貌似BGE后来的版本也把instruct取消了
感谢您耐心的解答
AiTutu changed discussion status to closed