Spaces:
Paused
Paused
File size: 3,701 Bytes
7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 2d44f1b 0eaf799 7dc9d46 2d44f1b 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 0eaf799 7dc9d46 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
import os
import gradio as gr
import openai
from dotenv import load_dotenv
# Load environment variables
load_dotenv()
openai.api_key = os.getenv("OPENAI_API_KEY")
# System prompt for intent/entity extraction
SYSTEM_PROMPT = (
"You are a customer support assistant. "
"For every user message, reply with a JSON object containing: "
"'intent' (the user's intent), 'entities' (dictionary of extracted entities), "
"and 'response' (your natural reply). "
'Example: {"intent": "order_status", "entities": {"order_number": "12345"}, "response": "Your order 12345 is being processed."} '
"If no entities, use an empty dictionary."
)
REQUEST_LIMIT = 200
global_request_count = 0
# Chatbot function using OpenAI ChatGPT API (OpenAI Python v1.x)
def chatbot(user_message, history):
global global_request_count
if global_request_count >= REQUEST_LIMIT:
display = "[Request limit reached. No further requests will be processed.]"
return history + [[user_message, display]]
global_request_count += 1
if history is None:
history = []
# Track known entities across conversation
known_entities = {}
import json
import re
# Extract entities from previous bot replies
for _, bot in history:
match = re.search(r"\{[\s\S]*\}", bot)
if match:
try:
parsed = json.loads(match.group(0))
entities = parsed.get("entities", {})
if isinstance(entities, dict):
known_entities.update(entities)
except Exception:
pass
# Prepend known entities to user message if any
if known_entities:
entity_context = f"[Known entities: {json.dumps(known_entities)}] "
user_message = entity_context + user_message
# Prepare messages for OpenAI API
messages = [{"role": "system", "content": SYSTEM_PROMPT}]
for user, bot in history:
messages.append({"role": "user", "content": user})
messages.append({"role": "assistant", "content": bot})
messages.append({"role": "user", "content": user_message})
try:
response = openai.chat.completions.create(
model="gpt-4.1-mini",
messages=messages,
max_completion_tokens=256,
temperature=0.7,
)
raw_content = response.choices[0].message.content.strip()
# Try to parse JSON
try:
# Extract JSON from code block or text
match = re.search(r"\{[\s\S]*\}", raw_content)
if match:
json_str = match.group(0)
parsed = json.loads(json_str)
display = (
f"**Intent:** {parsed.get('intent','')}\n"
f"**Entities:** {parsed.get('entities','')}\n"
f"**Bot:** {parsed.get('response','')}"
)
else:
display = f"[Could not parse intent/entities]\n{raw_content}"
except Exception:
display = f"[Could not parse intent/entities]\n{raw_content}"
except Exception as e:
display = f"[Error: {str(e)}]"
return history + [[user_message, display]]
with gr.Blocks() as demo:
gr.Markdown("# Customer Support Chatbot (MVP)")
chatbot_ui = gr.Chatbot()
msg = gr.Textbox(label="Your message")
clear = gr.Button("Clear")
def respond(user_message, chat_history):
new_history = chatbot(user_message, chat_history)
return "", new_history # Clear textbox after submit
msg.submit(respond, [msg, chatbot_ui], [msg, chatbot_ui])
clear.click(lambda: ("", None), None, [msg, chatbot_ui], queue=False)
demo.launch()
|