two version of R2R are here HEAD master

1 files changed, 31 insertions, 0 deletions

diff --git a/.venv/lib/python3.12/site-packages/shared/abstractions/embedding.py b/.venv/lib/python3.12/site-packages/shared/abstractions/embedding.py
new file mode 100644
index 00000000..6e27da28
--- /dev/null
+++ b/.venv/lib/python3.12/site-packages/shared/abstractions/embedding.py
@@ -0,0 +1,31 @@
+from enum import Enum, auto
+
+
+class EmbeddingPurpose(str, Enum):
+    INDEX = auto()
+    QUERY = auto()
+    DOCUMENT = auto()
+
+
+default_embedding_prefixes = {
+    "nomic-embed-text-v1.5": {
+        EmbeddingPurpose.INDEX: "",
+        EmbeddingPurpose.QUERY: "search_query: ",
+        EmbeddingPurpose.DOCUMENT: "search_document: ",
+    },
+    "nomic-embed-text": {
+        EmbeddingPurpose.INDEX: "",
+        EmbeddingPurpose.QUERY: "search_query: ",
+        EmbeddingPurpose.DOCUMENT: "search_document: ",
+    },
+    "mixedbread-ai/mxbai-embed-large-v1": {
+        EmbeddingPurpose.INDEX: "",
+        EmbeddingPurpose.QUERY: "Represent this sentence for searching relevant passages: ",
+        EmbeddingPurpose.DOCUMENT: "Represent this sentence for searching relevant passages: ",
+    },
+    "mixedbread-ai/mxbai-embed-large": {
+        EmbeddingPurpose.INDEX: "",
+        EmbeddingPurpose.QUERY: "Represent this sentence for searching relevant passages: ",
+        EmbeddingPurpose.DOCUMENT: "Represent this sentence for searching relevant passages: ",
+    },
+}