-
### Description
Some inference API providers now support embedding models with each dimension defined as a single bit. For example, the v3 models from Cohere offer this capability. Since we already h…
-
There seems to be an issue creating a field of type Collection(EdmSingle) in azure-search-documents==1.15.1 This could be a bug in azure-sdk-for-python
SearchFieldDataType.Collection(SearchFieldDat…
-
The extended descriptions of several vector diagnostics include the terms 'zonal' and 'meridional'. This is of unnecessary confusion when the model grid is not aligned with meridians and parallels (i…
-
### Description
The current implementation of retrievers follows closely to the original RRF (Reciprocal Rank Fusion) paper. I wonder if weighting could be introduced such that one result set has a h…
-
Hi, I'm trying to use dynamic snode to construct a vec3 field, but the documentation only gives the `ti.append()` method and only supports scalar inputs. Is there any way to construct a dynamic snode…
-
### Self Checks
- [X] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general).
- [X] I have s…
-
One may provide different ways to handle particles near the mesh boundary for example tangential or free flow (remove particle).
There has to be a GUI-element.
-
**Describe the bug**
`AzureAiSearchEmbeddingStoreIT` started failing, the error happens because `EmbeddingMatch.embedding()` returns `null`.
**Log and Stack trace**
```
Expected :Embedding { v…
-
code:
```
analyzer = build_default_analyzer(language="zh")
bm25_ef = BM25EmbeddingFunction(analyzer)
bm25_ef.load("D:/Downloads/bm25_msmarco_v1.json")
def test():
entities = [....]
for en…
-
**Is your feature request related to a problem? Please describe.**
One of the usage we want to have is research related. We have an entity, for which data scientists generate embeddings and then inse…