Update app.py
Browse files
app.py
CHANGED
|
@@ -30,39 +30,19 @@ def load_embeddings():
|
|
| 30 |
|
| 31 |
hf = load_embeddings()
|
| 32 |
|
| 33 |
-
@st.cache_data
|
| 34 |
-
def load_txt_documents(data_path):
|
| 35 |
-
documents = []
|
| 36 |
-
for filename in os.listdir(data_path):
|
| 37 |
-
if filename.endswith('.txt'):
|
| 38 |
-
file_path = os.path.join(data_path, filename)
|
| 39 |
-
documents.extend(TextLoader(file_path).load())
|
| 40 |
-
return documents
|
| 41 |
|
| 42 |
-
@st.cache_data
|
| 43 |
-
def load_uploaded_documents(uploaded_files):
|
| 44 |
-
documents = []
|
| 45 |
-
for uploaded_file in uploaded_files:
|
| 46 |
-
content = uploaded_file.read().decode("utf-8")
|
| 47 |
-
documents.append({"content": content, "filename": uploaded_file.name})
|
| 48 |
-
return documents
|
| 49 |
|
| 50 |
|
| 51 |
|
| 52 |
documents = load_documents(data_path)
|
| 53 |
docs = split_docs(documents, 450, 20)
|
| 54 |
|
| 55 |
-
@st.cache_resource
|
| 56 |
-
def create_chroma_db(docs, hf):
|
| 57 |
-
return Chroma(docs, hf)
|
| 58 |
|
| 59 |
-
chroma_db = create_chroma_db(docs, hf)
|
| 60 |
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
return mmr_retriever(chroma_db, "mmr", 6)
|
| 64 |
|
| 65 |
-
retriever =
|
| 66 |
|
| 67 |
# Set up LlamaCpp model
|
| 68 |
callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
|
|
|
|
| 30 |
|
| 31 |
hf = load_embeddings()
|
| 32 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 34 |
|
| 35 |
|
| 36 |
|
| 37 |
documents = load_documents(data_path)
|
| 38 |
docs = split_docs(documents, 450, 20)
|
| 39 |
|
|
|
|
|
|
|
|
|
|
| 40 |
|
|
|
|
| 41 |
|
| 42 |
+
chroma_db = chroma_db(docs, hf)
|
| 43 |
+
|
|
|
|
| 44 |
|
| 45 |
+
retriever = retriever_from_chroma(chroma_db,"mmr", 6)
|
| 46 |
|
| 47 |
# Set up LlamaCpp model
|
| 48 |
callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
|