Spaces:
Build error
Build error
Update ingest.py
Browse files
ingest.py
CHANGED
@@ -1,29 +1,29 @@
|
|
1 |
-
import os
|
2 |
-
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
3 |
-
from langchain.embeddings import SentenceTransformerEmbeddings
|
4 |
-
from langchain.document_loaders import DirectoryLoader
|
5 |
-
from langchain.document_loaders import PyPDFLoader
|
6 |
-
from langchain.vectorstores import Qdrant
|
7 |
-
from qdrant_client import QdrantClient
|
8 |
-
|
9 |
-
embeddings = SentenceTransformerEmbeddings(model_name="NeuML/pubmedbert-base-embeddings")
|
10 |
-
|
11 |
-
client = QdrantClient(
|
12 |
-
url=os.getenv("QDRANT_URL", "https://
|
13 |
-
api_key=os.getenv("QDRANT_API_KEY"),
|
14 |
-
prefer_grpc=False
|
15 |
-
)
|
16 |
-
|
17 |
-
loader = DirectoryLoader('data/', glob="**/*.pdf", show_progress=True, loader_cls=PyPDFLoader)
|
18 |
-
documents = loader.load()
|
19 |
-
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
|
20 |
-
texts = text_splitter.split_documents(documents)
|
21 |
-
|
22 |
-
qdrant = Qdrant.from_documents(
|
23 |
-
texts,
|
24 |
-
embeddings,
|
25 |
-
client=client,
|
26 |
-
collection_name="vector_db"
|
27 |
-
)
|
28 |
-
|
29 |
#print("Vector DB Successfully Created!")
|
|
|
1 |
+
import os
|
2 |
+
from langchain.text_splitter import RecursiveCharacterTextSplitter
|
3 |
+
from langchain.embeddings import SentenceTransformerEmbeddings
|
4 |
+
from langchain.document_loaders import DirectoryLoader
|
5 |
+
from langchain.document_loaders import PyPDFLoader
|
6 |
+
from langchain.vectorstores import Qdrant
|
7 |
+
from qdrant_client import QdrantClient
|
8 |
+
|
9 |
+
embeddings = SentenceTransformerEmbeddings(model_name="NeuML/pubmedbert-base-embeddings")
|
10 |
+
|
11 |
+
client = QdrantClient(
|
12 |
+
url=os.getenv("QDRANT_URL", "https://QDRANT_URL.europe-west3-0.gcp.cloud.qdrant.io"),
|
13 |
+
api_key=os.getenv("QDRANT_API_KEY"),
|
14 |
+
prefer_grpc=False
|
15 |
+
)
|
16 |
+
|
17 |
+
loader = DirectoryLoader('data/', glob="**/*.pdf", show_progress=True, loader_cls=PyPDFLoader)
|
18 |
+
documents = loader.load()
|
19 |
+
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
|
20 |
+
texts = text_splitter.split_documents(documents)
|
21 |
+
|
22 |
+
qdrant = Qdrant.from_documents(
|
23 |
+
texts,
|
24 |
+
embeddings,
|
25 |
+
client=client,
|
26 |
+
collection_name="vector_db"
|
27 |
+
)
|
28 |
+
|
29 |
#print("Vector DB Successfully Created!")
|