yujunjun/haystack
1
0
Fork 0
mirror of https://github.com/deepset-ai/haystack.git synced 2025-11-04 03:39:31 +00:00
Code Issues Packages Projects Releases Wiki Activity

Documenting output score of JoinDocuments when using concatenation (#2561)

Browse Source 
* add documentation regarding the score of JoinDocuments when using concatenation

* Update Documentation & Code Style

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
This commit is contained in:
MichelBartels 2022-05-16 18:30:07 +02:00 committed by GitHub
parent a2a99f79b1
commit 686e9d24ef
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
docs/_src/api/api/other.md
View File

@ -30,6 +30,7 @@ A node to join documents outputted by multiple retriever nodes.
The node allows multiple join modes:
* concatenate: combine the documents from multiple nodes. Any duplicate documents are discarded.
The score is only determined by the last node that outputs the document.
* merge: merge scores of documents from multiple nodes. Optionally, each input score can be given a different
`weight` & a `top_k` limit can be set. This mode can also be used for "reranking" retrieved documents.
* reciprocal_rank_fusion: combines the documents based on their rank in multiple nodes.

2
docs/_src/tutorials/tutorials/5.md
View File

@ -289,7 +289,7 @@ More info on this metric can be found in our [paper](https://arxiv.org/abs/2108.
```python
advanced_eval_result = pipeline.eval(
labels=eval_labels, params={"Retriever": {"top_k": 1}}, sas_model_name_or_path="cross-encoder/stsb-roberta-large"
labels=eval_labels, params={"Retriever": {"top_k": 5}}, sas_model_name_or_path="cross-encoder/stsb-roberta-large"
)
metrics = advanced_eval_result.calculate_metrics()

1
haystack/nodes/other/join_docs.py
View File

@ -12,6 +12,7 @@ class JoinDocuments(BaseComponent):
The node allows multiple join modes:
* concatenate: combine the documents from multiple nodes. Any duplicate documents are discarded.
The score is only determined by the last node that outputs the document.
* merge: merge scores of documents from multiple nodes. Optionally, each input score can be given a different
`weight` & a `top_k` limit can be set. This mode can also be used for "reranking" retrieved documents.
* reciprocal_rank_fusion: combines the documents based on their rank in multiple nodes.