From 0a6b7793e982d4c59f3c6ffb947dfe28ca709cdc Mon Sep 17 00:00:00 2001
From: Gustaf Rydholm <gustaf.rydholm@gmail.com>
Date: Fri, 5 Apr 2024 00:42:19 +0200
Subject: Rename

---
 rag/db/embeddings.py | 23 +++++++++++++++++++++++
 rag/db/vector.py     | 23 -----------------------
 2 files changed, 23 insertions(+), 23 deletions(-)
 create mode 100644 rag/db/embeddings.py
 delete mode 100644 rag/db/vector.py

(limited to 'rag/db')

diff --git a/rag/db/embeddings.py b/rag/db/embeddings.py
new file mode 100644
index 0000000..f9d0735
--- /dev/null
+++ b/rag/db/embeddings.py
@@ -0,0 +1,23 @@
+import os
+from typing import Tuple
+
+import faiss
+import numpy as np
+
+
+# TODO: inner product distance metric?
+class Embeddings:
+    def __init__(self):
+        self.dim = os.environ["EMBEDDING_DIM"]
+        self.index = faiss.IndexFlatL2(self.dim)
+        # TODO: load from file
+
+    def add(self, embeddings: np.ndarray):
+        # TODO: save to file
+        self.index.add(embeddings)
+
+    def search(
+        self, query: np.ndarray, neighbors: int = 4
+    ) -> Tuple[np.ndarray, np.ndarray]:
+        score, indices = self.index.search(query, neighbors)
+        return score, indices
diff --git a/rag/db/vector.py b/rag/db/vector.py
deleted file mode 100644
index f229ba7..0000000
--- a/rag/db/vector.py
+++ /dev/null
@@ -1,23 +0,0 @@
-from typing import Tuple
-import faiss
-import numpy as np
-
-# TODO: read from .env
-EMBEDDING_DIM = 1024
-
-
-# TODO: inner product distance metric?
-class VectorStore:
-    def __init__(self):
-        self.index = faiss.IndexFlatL2(EMBEDDING_DIM)
-        # TODO: load from file
-
-    def add(self, embeddings: np.ndarray):
-        # TODO: save to file
-        self.index.add(embeddings)
-
-    def search(
-        self, query: np.ndarray, neighbors: int = 4
-    ) -> Tuple[np.ndarray, np.ndarray]:
-        score, indices = self.index.search(query, neighbors)
-        return score, indices
-- 
cgit v1.2.3-70-g09d2