[Bug]: Qdrant fastembed sparse encoder model name typo

[manfredwang093]

Bug Description

There is a typo in the default sparse encoder model name

there is an 'i' missing from [prithivida/Splade_PP_en_v1](https://huggingface.co/prithivida/Splade_PP_en_v1)

llama_index/llama-index-integrations/vector_stores/llama-index-vector-stores-qdrant/llama_index/vector_stores/qdrant/utils.py

Line 68 in 58c00bd

model_name: str = "prithvida/Splade_PP_en_v1",

def fastembed_sparse_encoder(
model_name: str = "prithvida/Splade_PP_en_v1",
batch_size: int = 256,
cache_dir: Optional[str] = None,
threads: Optional[int] = None,
) -> SparseEncoderCallable:

Version

0.11.17

Steps to Reproduce

llama_index/llama-index-integrations/vector_stores/llama-index-vector-stores-qdrant/llama_index/vector_stores/qdrant/utils.py

Line 68 in 58c00bd

model_name: str = "prithvida/Splade_PP_en_v1",

Relevant Logs/Tracbacks

No response

[dosubot]

There is indeed a typo in the fastembed_sparse_encoder function. The model name should be "prithivida/Splade_PP_en_v1" instead of "prithvida/Splade_PP_en_v1" ^[1].

To continue talking to Dosu, mention @dosu.

[logan-markewich]

Feel free to make a PR!

[manfredwang093]

Thanks Logan for your quick response, tried to push my branch, but got message saying "you don't have write access to run-llama/llama_index", is there a better way for me to create a PR, or how may I request write access? Thanks.

In a separate thread regarding LanceDB, there is a bug I shared there earlier regarding issues in data ingestion process. Could you take a look when have time? Thank you.
#14435 (comment)

[logan-markewich]

@manfredwang093 normally you need to fork the repo, push changes to your fork, and then propose a PR with your fork

[manfredwang093]

@manfredwang093 normally you need to fork the repo, push changes to your fork, and then propose a PR with your fork

Thank you @logan-markewich , just opened this PR
#16494