Spaces:
Running
Running
deploy: rename Dockerfile.hf -> Dockerfile on the Space side
Browse files- Dockerfile +8 -4
Dockerfile
CHANGED
|
@@ -104,10 +104,14 @@ ENV SAR_LOCAL_EMBEDDING_MODEL=BAAI/bge-m3
|
|
| 104 |
ENV SAR_EMBEDDING_MODEL=bge-m3
|
| 105 |
ENV SAR_EMBEDDING_DIM=1024
|
| 106 |
|
| 107 |
-
#
|
| 108 |
-
#
|
| 109 |
-
#
|
| 110 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 111 |
ENV SAR_RERANKER_CHECKPOINT=BAAI/bge-reranker-v2-m3
|
| 112 |
|
| 113 |
# Sparse retrieval -- BM25 keeps the cold path zero-dep; SPLADE adds an
|
|
|
|
| 104 |
ENV SAR_EMBEDDING_MODEL=bge-m3
|
| 105 |
ENV SAR_EMBEDDING_DIM=1024
|
| 106 |
|
| 107 |
+
# Reranker disabled for the BYOK demo. The cross-encoder adds a ~600 MB
|
| 108 |
+
# model + 4-5 s cold-load latency, and on a 10-doc corpus + 1-5 visitor
|
| 109 |
+
# uploads its top-5 cut routinely drops the visitor's own chunk before
|
| 110 |
+
# it reaches the grader. Pure dense+sparse RRF order is good enough on
|
| 111 |
+
# the demo corpus; bench shows the reranker only helps materially past
|
| 112 |
+
# 200+ documents per query. Re-enable for production deploys with
|
| 113 |
+
# SAR_RERANKER_TYPE=cross_encoder or fine_tuned.
|
| 114 |
+
ENV SAR_RERANKER_TYPE=none
|
| 115 |
ENV SAR_RERANKER_CHECKPOINT=BAAI/bge-reranker-v2-m3
|
| 116 |
|
| 117 |
# Sparse retrieval -- BM25 keeps the cold path zero-dep; SPLADE adds an
|