Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -8,7 +8,6 @@ from llama_index.llms.openai import OpenAI
|
|
8 |
from llama_index.embeddings.openai import OpenAIEmbedding
|
9 |
from llama_index.vector_stores.faiss import FaissVectorStore
|
10 |
from llama_index.core.ingestion import IngestionPipeline
|
11 |
-
from langchain_community.document_loaders.csv_loader import CSVLoader
|
12 |
from langchain_community.vectorstores import FAISS as LangChainFAISS
|
13 |
from langchain.chains import create_retrieval_chain
|
14 |
from langchain.chains.combine_documents import create_stuff_documents_chain
|
@@ -42,9 +41,9 @@ if uploaded_file:
|
|
42 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".csv", mode="w", encoding="utf-8") as temp_file:
|
43 |
temp_file_path = temp_file.name
|
44 |
data.to_csv(temp_file.name, index=False, encoding="utf-8")
|
45 |
-
temp_file.flush()
|
46 |
|
47 |
-
#
|
48 |
st.write("Temporary file path:", temp_file_path)
|
49 |
with open(temp_file_path, "r") as f:
|
50 |
st.write("Temporary file content:")
|
@@ -53,19 +52,21 @@ if uploaded_file:
|
|
53 |
# Tabs for LangChain and LlamaIndex
|
54 |
tab1, tab2 = st.tabs(["LangChain", "LlamaIndex"])
|
55 |
|
56 |
-
# LangChain Tab
|
57 |
with tab1:
|
58 |
st.subheader("LangChain Query")
|
59 |
try:
|
60 |
-
#
|
61 |
-
st.write("
|
62 |
-
|
63 |
-
|
|
|
|
|
64 |
|
65 |
# Debugging: Preview loaded documents
|
66 |
-
st.write("Successfully
|
67 |
-
if
|
68 |
-
st.text(
|
69 |
|
70 |
# Create FAISS VectorStore
|
71 |
langchain_index = faiss.IndexFlatL2(EMBED_DIMENSION)
|
@@ -73,7 +74,7 @@ if uploaded_file:
|
|
73 |
embedding_function=OpenAIEmbeddings(),
|
74 |
index=langchain_index,
|
75 |
)
|
76 |
-
langchain_vector_store.add_documents(
|
77 |
|
78 |
# LangChain Retrieval Chain
|
79 |
retriever = langchain_vector_store.as_retriever()
|
|
|
8 |
from llama_index.embeddings.openai import OpenAIEmbedding
|
9 |
from llama_index.vector_stores.faiss import FaissVectorStore
|
10 |
from llama_index.core.ingestion import IngestionPipeline
|
|
|
11 |
from langchain_community.vectorstores import FAISS as LangChainFAISS
|
12 |
from langchain.chains import create_retrieval_chain
|
13 |
from langchain.chains.combine_documents import create_stuff_documents_chain
|
|
|
41 |
with tempfile.NamedTemporaryFile(delete=False, suffix=".csv", mode="w", encoding="utf-8") as temp_file:
|
42 |
temp_file_path = temp_file.name
|
43 |
data.to_csv(temp_file.name, index=False, encoding="utf-8")
|
44 |
+
temp_file.flush()
|
45 |
|
46 |
+
# Verify the temporary file
|
47 |
st.write("Temporary file path:", temp_file_path)
|
48 |
with open(temp_file_path, "r") as f:
|
49 |
st.write("Temporary file content:")
|
|
|
52 |
# Tabs for LangChain and LlamaIndex
|
53 |
tab1, tab2 = st.tabs(["LangChain", "LlamaIndex"])
|
54 |
|
55 |
+
# LangChain Tab with Custom Loader
|
56 |
with tab1:
|
57 |
st.subheader("LangChain Query")
|
58 |
try:
|
59 |
+
# Custom preprocessing: Split CSV rows into smaller chunks
|
60 |
+
st.write("Processing CSV with a custom loader...")
|
61 |
+
documents = []
|
62 |
+
for _, row in data.iterrows():
|
63 |
+
content = "\n".join([f"{col}: {row[col]}" for col in data.columns])
|
64 |
+
documents.append({"page_content": content})
|
65 |
|
66 |
# Debugging: Preview loaded documents
|
67 |
+
st.write("Successfully processed documents:")
|
68 |
+
if documents:
|
69 |
+
st.text(documents[0]["page_content"])
|
70 |
|
71 |
# Create FAISS VectorStore
|
72 |
langchain_index = faiss.IndexFlatL2(EMBED_DIMENSION)
|
|
|
74 |
embedding_function=OpenAIEmbeddings(),
|
75 |
index=langchain_index,
|
76 |
)
|
77 |
+
langchain_vector_store.add_documents(documents)
|
78 |
|
79 |
# LangChain Retrieval Chain
|
80 |
retriever = langchain_vector_store.as_retriever()
|