FlagEmbedding
FlagEmbedding copied to clipboard
Retrieval and Retrieval-augmented LLMs
In the split_data_by_length.py code inside BGE-M3, after filtering the dataset by "max_length" field, the "idx" field is somehow changed , so the `split_dataset = dataset.select(idxs["idx"])` will result in the wrong...
加载模型的时候报错:
from FlagEmbedding import BGEM3FlagModel 通过这种方式进行部署推理的,但是压测的时候,发现CPU利用率达到200%+, GPU利用率仅仅2%,T4的卡,有啥建议? [INFO ] WorkerPool - loading model bge_m3_deploy_code (PENDING) on gpu(0) ... -- [INFO ] ModelInfo - S3 url found, start downloading from s3://sagemaker-us-west-2-106839800180/LLM-RAG/workshop/bge-m3-model/ [INFO...
微调参数设置
你好,我有以下疑问: 1. 微调时,per_device_train_batch_size设置为多少比较合适呢? 2. 若采用use_inbatch_neg,batch_size是否越大越好呢?
return_sparse_embedding arg exists in BGEM3ForInference, but cannot pass this arg from BGEM3FlagModel
Not sure how to generate Sparse Embeddings when using langchain.
请问如何使用langchain获取bge-m3的稀疏向量分数
- 问题1:两个一样的句子,其分数也没有到达10  - 问题2: 分数的分布的区间是什么呢,如果想归一化,有推荐的归一化方法或函数么
bge-m3中3种混合检索的方法用什么向量数据库可以支持呢?
Same sentences can always get a "1" simirlar score like dense way but not a score less than 1 and change with different sentence content. Different sentences can get an...