Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -52,20 +52,23 @@ def split_text(documents: list[Document]):
|
|
52 |
|
53 |
|
54 |
def save_to_chroma(chunks: list[Document]):
|
55 |
-
# Clear out the database first
|
56 |
if os.path.exists(CHROMA_PATH):
|
57 |
shutil.rmtree(CHROMA_PATH)
|
58 |
|
59 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
60 |
db = Chroma.from_documents(
|
61 |
chunks,
|
62 |
-
|
63 |
-
cache_folder="model_cache", # Укажите путь для кэша
|
64 |
persist_directory=CHROMA_PATH
|
65 |
)
|
66 |
-
|
67 |
print(f"Saved {len(chunks)} chunks to {CHROMA_PATH}.")
|
68 |
|
69 |
-
|
70 |
if __name__ == "__main__":
|
71 |
main()
|
|
|
52 |
|
53 |
|
54 |
def save_to_chroma(chunks: list[Document]):
|
55 |
+
# Clear out the database first
|
56 |
if os.path.exists(CHROMA_PATH):
|
57 |
shutil.rmtree(CHROMA_PATH)
|
58 |
|
59 |
+
# Initialize embeddings with cache
|
60 |
+
embeddings = HuggingFaceEmbeddings(
|
61 |
+
model_name="BAAI/bge-m3",
|
62 |
+
cache_folder="model_cache" # Правильное место для кэша
|
63 |
+
)
|
64 |
+
|
65 |
+
# Create Chroma DB
|
66 |
db = Chroma.from_documents(
|
67 |
chunks,
|
68 |
+
embeddings, # Используем предварительно созданный объект
|
|
|
69 |
persist_directory=CHROMA_PATH
|
70 |
)
|
|
|
71 |
print(f"Saved {len(chunks)} chunks to {CHROMA_PATH}.")
|
72 |
|
|
|
73 |
if __name__ == "__main__":
|
74 |
main()
|