-
In FlagEmbedding, hard negative mining is extracted based on ranking. (FlagEmbedding/baai_general_embedding/finetune/hn_mine.py) Is there a code that does hard negative mining based on similarity scor…
-
**Is your feature request related to a problem? Please describe.**
I'm trying to be able to add a thresold to my similarity search with vectors but it doesn't semm to be mentioned here anywhere?
ht…
-
Hello,
First of all, congratulations on your paper! I found it insightful and truly enjoyed reading through it.
I was wondering if there is a Python version of your approach. Specifically, for a…
-
My project uses RAFT brute force knn and I noticed a drop in precision when I upgraded to the latest RAFT. I moved to cuvs but still see cosine similarity that's off by more than 1e-04 when compared t…
phact updated
1 month ago
-
[] Truncated SVD is used for dimensionality reduction in order to account for scaling -> right now the ratio is 1/5. Is this optimal? What are things to consider for the future
[] Titles are duplicat…
-
Related: https://github.com/pytorch/pytorch/pull/31378
Current implementation: https://github.com/rusty1s/pytorch_cluster/blob/dbcafbe6a60aaa631b39050e3aa228f6d3fd1592/csrc/cuda/knn_cuda.cu#L56
As…
-
https://github.com/scikit-learn/scikit-learn/blob/0fb307bf39bbdacd6ed713c00724f8f871d60370/sklearn/metrics/pairwise.py#L1144
reproduce the greater than 1.0 case:
```python
from sklearn.metrics.pair…
wwz58 updated
5 months ago
-
Hello Team,
Thank you for your amazing work on this model. I was able to reproduce your remarkable results. I am looking to contribute and develop downstream inference using faiss but I am running …
-
Does anyone have an example python script that uses one on the x-vector extraction models developed here to extract embeddings? I've gone through some of the repo and have not found any such thing.
…
-
I wanted to know if I could be provided some examples to understand how we can create a vectorDB using embeddings and correspodning url. And index it using Python. We have memory constraints like we h…