Update app.py

app.py

@@ -7,7 +7,6 @@ For more information on `huggingface_hub` Inference API support, please check th
 """
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
-
 def extract_text_from_pdf(pdf_path):
     # Open the provided PDF file
     doc = fitz.open(pdf_path)
@@ -17,9 +16,9 @@ def extract_text_from_pdf(pdf_path):
     for page in doc:
         text += page.get_text()
 
+    doc.close()  # Ensure the PDF file is closed
     return text
 
-
 def respond(
     message,
     history: list[tuple[str, str]],
@@ -50,9 +49,9 @@ def respond(
         token = message.choices[0].delta.content
 
         response += token
+        print(f"Token: {token}")  # Debugging statement to trace tokens
         yield response
 
-
 def process_resume_and_respond(pdf_file, message, history, system_message, max_tokens, temperature, top_p):
     # Extract text from the PDF file
     resume_text = extract_text_from_pdf(pdf_file.name)
@@ -63,7 +62,6 @@ def process_resume_and_respond(pdf_file, message, history, system_message, max_t
     response = "".join([token for token in response_gen])
     return response
 
-
 # Store the uploaded PDF content globally
 uploaded_resume_text = ""
 
@@ -72,7 +70,6 @@ def upload_resume(pdf_file):
     uploaded_resume_text = extract_text_from_pdf(pdf_file.name)
     return "Resume uploaded successfully!"
 
-
 def respond_with_resume(message, history, system_message, max_tokens, temperature, top_p):
     global uploaded_resume_text
     # Combine the uploaded resume text with the user message
@@ -82,7 +79,6 @@ def respond_with_resume(message, history, system_message, max_tokens, temperatur
     response = "".join([token for token in response_gen])
     return response
 
-
 """
 For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
 """
@@ -113,6 +109,5 @@ demo = gr.TabbedInterface(
     ["Upload Resume", "Chat with Job Advisor"]
 )
 
-
 if __name__ == "__main__":
     demo.launch()