Spaces:
Sleeping
Sleeping
Update app.py
Browse filesupdated the code to handle no pdf file provided
app.py
CHANGED
|
@@ -73,20 +73,27 @@ def split_text(text, chunk_size=500):
|
|
| 73 |
def chatbot(pdf_file, user_question):
|
| 74 |
"""Processes the PDF and answers the user's question."""
|
| 75 |
print("chatbot start")
|
| 76 |
-
|
| 77 |
-
# Extract text from the PDF
|
| 78 |
-
text = extract_text_from_pdf(pdf_file)
|
| 79 |
-
if not text:
|
| 80 |
-
return "Could not extract any text from the PDF."
|
| 81 |
-
|
| 82 |
-
# retrieve the document relevant to the query
|
| 83 |
-
doc = retrieve_document(user_question)
|
| 84 |
|
| 85 |
-
|
| 86 |
-
|
|
|
|
|
|
|
|
|
|
| 87 |
|
| 88 |
-
|
| 89 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
|
| 91 |
try:
|
| 92 |
response = together.Completion.create(
|
|
|
|
| 73 |
def chatbot(pdf_file, user_question):
|
| 74 |
"""Processes the PDF and answers the user's question."""
|
| 75 |
print("chatbot start")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 76 |
|
| 77 |
+
if pdf_file:
|
| 78 |
+
# Extract text from the PDF
|
| 79 |
+
text = extract_text_from_pdf(pdf_file)
|
| 80 |
+
if not text:
|
| 81 |
+
return "Could not extract any text from the PDF."
|
| 82 |
|
| 83 |
+
try:
|
| 84 |
+
# retrieve the document relevant to the query
|
| 85 |
+
doc = retrieve_document(user_question)
|
| 86 |
+
except Exception as e:
|
| 87 |
+
return f"Error retrieving document relevant to the query: {user_question} \n{e}"
|
| 88 |
+
|
| 89 |
+
if doc:
|
| 90 |
+
# Split into smaller chunks
|
| 91 |
+
chunks = split_text(doc)
|
| 92 |
+
|
| 93 |
+
# Use only the first chunk (to optimize token usage)
|
| 94 |
+
prompt = f"Based on this document, answer the question:\n\nDocument:\n{chunks[0]}\n\nQuestion: {user_question}"
|
| 95 |
+
else:
|
| 96 |
+
prompt=user_question
|
| 97 |
|
| 98 |
try:
|
| 99 |
response = together.Completion.create(
|