QuaidKhalid commited on
Commit
aa1686a
·
verified ·
1 Parent(s): 99c3c54

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +13 -5
app.py CHANGED
@@ -7,6 +7,8 @@ from langchain_core.runnables import RunnablePassthrough
7
  from langchain_text_splitters import RecursiveCharacterTextSplitter
8
  from sentence_transformers import SentenceTransformer
9
  import torch
 
 
10
 
11
 
12
  # Define the embedding class
@@ -35,11 +37,15 @@ langchain_api_key = "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"
35
 
36
  def load_document(document_path):
37
  try:
38
- loader = PyPDFLoader(document_path)
39
- docs = loader.load()
40
- text_splitter = RecursiveCharacterTextSplitter(chunk_size=4000, chunk_overlap=200)
41
- splits = text_splitter.split_documents(docs)
42
- return splits
 
 
 
 
43
  except Exception as e:
44
  return str(e)
45
 
@@ -85,4 +91,6 @@ if document_path is not None and query:
85
  result = answer_question(rag_chain, query)
86
  st.write("Result:", result)
87
 
 
 
88
  st.write("Note: Replace `llm` with an appropriate language model.")
 
7
  from langchain_text_splitters import RecursiveCharacterTextSplitter
8
  from sentence_transformers import SentenceTransformer
9
  import torch
10
+ import os
11
+ import tempfile
12
 
13
 
14
  # Define the embedding class
 
37
 
38
  def load_document(document_path):
39
  try:
40
+ with tempfile.TemporaryDirectory() as tmp_dir:
41
+ tmp_file = os.path.join(tmp_dir, 'temp.pdf')
42
+ with open(tmp_file, 'wb') as f:
43
+ f.write(document_path.getvalue())
44
+ loader = PyPDFLoader(tmp_file)
45
+ docs = loader.load()
46
+ text_splitter = RecursiveCharacterTextSplitter(chunk_size=4000, chunk_overlap=200)
47
+ splits = text_splitter.split_documents(docs)
48
+ return splits
49
  except Exception as e:
50
  return str(e)
51
 
 
91
  result = answer_question(rag_chain, query)
92
  st.write("Result:", result)
93
 
94
+
95
+
96
  st.write("Note: Replace `llm` with an appropriate language model.")