diff --git a/semantic_router/index/base.py b/semantic_router/index/base.py
index c3c9ed0fb740e7195e723965a23dddedeb2948bc..77053ef08100a5d7dd5f06048ce9360b0987f091 100644
--- a/semantic_router/index/base.py
+++ b/semantic_router/index/base.py
@@ -13,8 +13,8 @@ class BaseIndex(BaseModel):
# You can define common attributes here if there are any.
# For example, a placeholder for the index attribute:
index: Optional[Any] = None
- routes: Optional[List[str]] = None
- utterances: Optional[List[str]] = None
+ routes: Optional[np.ndarray] = None
+ utterances: Optional[np.ndarray] = None
dimensions: Union[int, None] = None
type: str = "base"
diff --git a/semantic_router/index/local.py b/semantic_router/index/local.py
index 08cd088591850f2b5da2aefa5a430fbf7870cbb2..058ee2bc08e562a56327c61850471b14409dfa6a 100644
--- a/semantic_router/index/local.py
+++ b/semantic_router/index/local.py
@@ -8,12 +8,10 @@ class LocalIndex(BaseIndex):
def __init__(
self,
index: Optional[np.ndarray] = None,
- routes: Optional[List[str]] = None,
- utterances: Optional[List[str]] = None,
+ routes: Optional[np.ndarray] = None,
+ utterances: Optional[np.ndarray] = None,
):
- super().__init__(
- index=index, routes=routes, utterances=utterances
- )
+ super().__init__(index=index, routes=routes, utterances=utterances)
self.type = "local"
class Config: # Stop pydantic from complaining about Optional[np.ndarray] type hints.
@@ -41,9 +39,9 @@ class LocalIndex(BaseIndex):
Returns:
List[Tuple]: A list of (route_name, utterance) objects.
"""
- if self.route_names is None or self.utterances is None:
+ if self.routes is None or self.utterances is None:
raise ValueError("No routes have been added to the index.")
- return list(zip(self.route_names, self.utterances))
+ return list(zip(self.routes, self.utterances))
def describe(self) -> dict:
return {
@@ -64,7 +62,7 @@ class LocalIndex(BaseIndex):
# get routes from index values
route_names = self.routes[idx].copy()
return scores, route_names
-
+
def delete(self, route_name: str):
"""
Delete all records of a specific route from the index.
@@ -95,7 +93,7 @@ class LocalIndex(BaseIndex):
raise ValueError("Routes are not populated.")
idx = [i for i, route in enumerate(self.routes) if route == route_name]
return idx
-
+
def __len__(self):
if self.index is not None:
return self.index.shape[0]
diff --git a/semantic_router/index/pinecone.py b/semantic_router/index/pinecone.py
index 4594962f5a3d612dadcf7161c8bb6001d5d1edce..a6492226a46ebd2550f9c467fa64aa44af36d0a3 100644
--- a/semantic_router/index/pinecone.py
+++ b/semantic_router/index/pinecone.py
@@ -3,7 +3,7 @@ import requests
import time
import hashlib
import os
-from typing import Any, List, Tuple, Optional, Union
+from typing import Any, Dict, List, Tuple, Optional, Union
from semantic_router.index.base import BaseIndex
from semantic_router.utils.logger import logger
import numpy as np
@@ -128,11 +128,13 @@ class PineconeIndex(BaseIndex):
clean_route = clean_route_name(route_name)
ids, _ = self._get_all(prefix=f"{clean_route}#")
return ids
-
+
def _get_all(self, prefix: Optional[str] = None, include_metadata: bool = False):
"""
Retrieves all vector IDs from the Pinecone index using pagination.
"""
+ if self.index is None:
+ raise ValueError("Index is None, could not retrieve vector IDs.")
all_vector_ids = []
next_page_token = None
@@ -143,8 +145,8 @@ class PineconeIndex(BaseIndex):
# Construct the request URL for listing vectors. Adjust parameters as needed.
list_url = f"https://{self.host}/vectors/list{prefix_str}"
- params = {}
- headers = {"Api-Key": os.getenv("PINECONE_API_KEY")}
+ params: Dict = {}
+ headers = {"Api-Key": os.environ["PINECONE_API_KEY"]}
metadata = []
while True:
@@ -164,9 +166,7 @@ class PineconeIndex(BaseIndex):
if include_metadata:
res_meta = self.index.fetch(ids=vector_ids)
# extract metadata only
- metadata.extend(
- [x["metadata"] for x in res_meta["vectors"].values()]
- )
+ metadata.extend([x["metadata"] for x in res_meta["vectors"].values()])
# Check if there's a next page token; if not, break the loop
next_page_token = response_data.get("pagination", {}).get("next")
@@ -174,7 +174,7 @@ class PineconeIndex(BaseIndex):
break
return all_vector_ids, metadata
-
+
def get_routes(self) -> List[Tuple]:
"""
Gets a list of route and utterance objects currently stored in the index.
diff --git a/semantic_router/layer.py b/semantic_router/layer.py
index bec876748caa563f2941bb7d800c240466d9930b..5d56aefb96f86881838b960c9c80f02b28f40f29 100644
--- a/semantic_router/layer.py
+++ b/semantic_router/layer.py
@@ -303,8 +303,7 @@ class RouteLayer:
self.index.delete(route_name=route_name)
def _refresh_routes(self):
- """Pulls out the latest routes from the index.
- """
+ """Pulls out the latest routes from the index."""
raise NotImplementedError("This method has not yet been implemented.")
route_mapping = {route.name: route for route in self.routes}
index_routes = self.index.get_routes()
@@ -320,7 +319,6 @@ class RouteLayer:
route = route_mapping[route_name]
self.routes.append(route)
-
def _add_routes(self, routes: List[Route]):
# create embeddings for all routes
all_utterances = [
@@ -458,7 +456,7 @@ class RouteLayer:
correct += 1
accuracy = correct / len(Xq)
return accuracy
-
+
def _get_route_names(self) -> List[str]:
return [route.name for route in self.routes]
diff --git a/tests/unit/test_layer.py b/tests/unit/test_layer.py
index e5ba76a2f2fbd997cd1db9f1d593aa6e9a834cbe..03c0c1eaba540ec6f0f9ae0dc93ab38e41a3265d 100644
--- a/tests/unit/test_layer.py
+++ b/tests/unit/test_layer.py
@@ -302,7 +302,9 @@ class TestRouteLayer:
# now load from config and confirm it's the same
route_layer_from_config = RouteLayer.from_config(layer_config)
assert (route_layer_from_config.index.index == route_layer.index.index).all()
- assert route_layer_from_config._get_route_names() == route_layer._get_route_names()
+ assert (
+ route_layer_from_config._get_route_names() == route_layer._get_route_names()
+ )
assert route_layer_from_config.score_threshold == route_layer.score_threshold
def test_get_thresholds(self, openai_encoder, routes):