Spaces:
Sleeping
Sleeping
Palbha Kulkarni (Nazwale)
commited on
Commit
·
4460c42
1
Parent(s):
bd044ff
Created using Colab
Browse files- faq_data_generator.ipynb +232 -0
faq_data_generator.ipynb
ADDED
@@ -0,0 +1,232 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"nbformat": 4,
|
3 |
+
"nbformat_minor": 0,
|
4 |
+
"metadata": {
|
5 |
+
"colab": {
|
6 |
+
"provenance": [],
|
7 |
+
"authorship_tag": "ABX9TyNAxD9Hy7SaN4kD/p7d0PC5",
|
8 |
+
"include_colab_link": true
|
9 |
+
},
|
10 |
+
"kernelspec": {
|
11 |
+
"name": "python3",
|
12 |
+
"display_name": "Python 3"
|
13 |
+
},
|
14 |
+
"language_info": {
|
15 |
+
"name": "python"
|
16 |
+
}
|
17 |
+
},
|
18 |
+
"cells": [
|
19 |
+
{
|
20 |
+
"cell_type": "markdown",
|
21 |
+
"metadata": {
|
22 |
+
"id": "view-in-github",
|
23 |
+
"colab_type": "text"
|
24 |
+
},
|
25 |
+
"source": [
|
26 |
+
"<a href=\"https://colab.research.google.com/github/palbha/airline-faq-rag/blob/main/faq_data_generator.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
|
27 |
+
]
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"cell_type": "markdown",
|
31 |
+
"source": [
|
32 |
+
"# Install necessary libraries if required - This code ran on Google Colab & the libraires where supported by default- please rephrase this\n",
|
33 |
+
"\n"
|
34 |
+
],
|
35 |
+
"metadata": {
|
36 |
+
"id": "-Ruq0mXsA9do"
|
37 |
+
}
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"cell_type": "code",
|
41 |
+
"source": [
|
42 |
+
"import csv\n",
|
43 |
+
"import json\n",
|
44 |
+
"import os\n",
|
45 |
+
"import openai\n",
|
46 |
+
"import csv\n",
|
47 |
+
"from openai import OpenAI\n",
|
48 |
+
"from google.colab import userdata\n",
|
49 |
+
"\n",
|
50 |
+
"#Based on airlines FAQ I identified potential topics which can be shared with our agent to create FAQ's\n",
|
51 |
+
"FAQ_TOPICS = [\n",
|
52 |
+
" \"Airport Services\",\n",
|
53 |
+
" \"Animal Transportation\",\n",
|
54 |
+
" \"Beyond Business\",\n",
|
55 |
+
" \"Booking and managing a reservation\",\n",
|
56 |
+
" \"Carbon Offsetting\",\n",
|
57 |
+
" \"AirlineX Compliance\",\n",
|
58 |
+
" \"Hotels, cars and travel insurance\",\n",
|
59 |
+
" \"AirlineX Offers\",\n",
|
60 |
+
" \"On-board experience\",\n",
|
61 |
+
" \"Operational Updates\",\n",
|
62 |
+
" \"Payments\",\n",
|
63 |
+
" \"Privilege Club : Qatar Airways' loyalty programme\",\n",
|
64 |
+
" \"ArlineX Airways Affiliate Program\",\n",
|
65 |
+
" \"ArlineX Airways Packages\",\n",
|
66 |
+
" \"ATravel - ArlineX Loyalty Program\",\n",
|
67 |
+
" \"ATravel - ArlineX Loyalty Program - Account Cancellation\",\n",
|
68 |
+
" \"ATravel - ArlineX Loyalty Program - Account Management\",\n",
|
69 |
+
" \"ATravel - ArlineX Loyalty Program - Booking Terms and Conditions\",\n",
|
70 |
+
" \"Travel Baggage\",\n",
|
71 |
+
" \"Baggage\",\n",
|
72 |
+
" \"BAGTAG\",\n",
|
73 |
+
" \"Hand baggage\",\n",
|
74 |
+
" \"Liquids\",\n",
|
75 |
+
" \"Mishandled baggage\",\n",
|
76 |
+
" \"Travel voucher\",\n",
|
77 |
+
" \"Voucher redemption\",\n",
|
78 |
+
" \"TripAdd\",\n",
|
79 |
+
" \"eSIM - TripAdd\",\n",
|
80 |
+
" \"Lounge - TripAdd\",\n",
|
81 |
+
" \"Meet and Greet - TripAdd\",\n",
|
82 |
+
" \"Young Travellers\",\n",
|
83 |
+
" \"Travelling with children\",\n",
|
84 |
+
" \"Unaccompanied minors\",\n",
|
85 |
+
"]"
|
86 |
+
],
|
87 |
+
"metadata": {
|
88 |
+
"id": "aAvEjDrrRNJE"
|
89 |
+
},
|
90 |
+
"execution_count": 2,
|
91 |
+
"outputs": []
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"cell_type": "code",
|
95 |
+
"source": [
|
96 |
+
"openai = OpenAI(\n",
|
97 |
+
" base_url=\"https://generativelanguage.googleapis.com/v1beta/\",\n",
|
98 |
+
" api_key=userdata.get('gemini_api'),\n",
|
99 |
+
")"
|
100 |
+
],
|
101 |
+
"metadata": {
|
102 |
+
"id": "SWnjOAg7BeO_"
|
103 |
+
},
|
104 |
+
"execution_count": 3,
|
105 |
+
"outputs": []
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"cell_type": "code",
|
109 |
+
"source": [
|
110 |
+
"def get_faq_ques_for_topic(topic):\n",
|
111 |
+
" messages = [\n",
|
112 |
+
" {\n",
|
113 |
+
" \"role\": \"system\",\n",
|
114 |
+
" \"content\": (\n",
|
115 |
+
" \"\"\"You are an assistant that generates FAQ-style questions for an airline named Airline X, which operates international and domestic flights in Canada.\n",
|
116 |
+
"\n",
|
117 |
+
"For each topic, Generate realistic and informative user-style questions for the FAQ topic. Do no include answers\n",
|
118 |
+
"\"\"\"\n",
|
119 |
+
" )\n",
|
120 |
+
" },\n",
|
121 |
+
" {\n",
|
122 |
+
" \"role\": \"user\",\n",
|
123 |
+
" \"content\": f\"Generate FAQ 5-10 questions about the topic: '{topic}'.\"\n",
|
124 |
+
" },\n",
|
125 |
+
" {\n",
|
126 |
+
" \"role\": \"system\",\n",
|
127 |
+
" \"content\": \"Return ONLY a valid JSON array of question objects. Do not include answers\"\n",
|
128 |
+
" }\n",
|
129 |
+
" ]\n",
|
130 |
+
" response = openai.chat.completions.create(\n",
|
131 |
+
" model=\"gemini-1.5-flash\",\n",
|
132 |
+
" messages=messages,\n",
|
133 |
+
" temperature=0.7,\n",
|
134 |
+
" max_tokens=700,\n",
|
135 |
+
" )\n",
|
136 |
+
" # The model response should be a JSON array of objects like: [{\"question\": \"...\", \"answer\": \"...\"}, ...]\n",
|
137 |
+
" content = response.choices[0].message.content.strip()\n",
|
138 |
+
"\n",
|
139 |
+
"\n",
|
140 |
+
" return content\n",
|
141 |
+
"\n"
|
142 |
+
],
|
143 |
+
"metadata": {
|
144 |
+
"id": "64J-8B5hRSMf"
|
145 |
+
},
|
146 |
+
"execution_count": 4,
|
147 |
+
"outputs": []
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"cell_type": "code",
|
151 |
+
"source": [
|
152 |
+
"def get_faq_ans_for_topic(topic,question):\n",
|
153 |
+
" messages = [\n",
|
154 |
+
" {\n",
|
155 |
+
" \"role\": \"system\",\n",
|
156 |
+
" \"content\": (\n",
|
157 |
+
" f\"You are an assistant that generates FAQ-style answers for an airline named Airline X, \"\n",
|
158 |
+
" f\"which operates international and domestic flights in Canada.\\n\\n\"\n",
|
159 |
+
" f\"The FAQ topic is: '{topic}'.\\n\"\n",
|
160 |
+
"\n",
|
161 |
+
" )\n",
|
162 |
+
" },\n",
|
163 |
+
" {\n",
|
164 |
+
" \"role\": \"user\",\n",
|
165 |
+
" \"content\": f\"Generate FAQ answers for the question: '{question}'.\"\n",
|
166 |
+
" } ,\n",
|
167 |
+
" {\n",
|
168 |
+
" \"role\": \"system\",\n",
|
169 |
+
" \"content\": \"Provide a clear, self-contained, and factual-sounding answer based on Airline X's own policies. \"\n",
|
170 |
+
" \"Do NOT reference any website, support, or external links. Make the answer complete, realistic, \"\n",
|
171 |
+
" \"and independent of outside context. Return ONLY answers\"\n",
|
172 |
+
" }\n",
|
173 |
+
" ]\n",
|
174 |
+
" response = openai.chat.completions.create(\n",
|
175 |
+
" model=\"gemini-1.5-flash\",\n",
|
176 |
+
" messages=messages,\n",
|
177 |
+
" temperature=0.7,\n",
|
178 |
+
" max_tokens=700,\n",
|
179 |
+
" )\n",
|
180 |
+
" # The model response should be a JSON array of objects like: [{\"question\": \"...\", \"answer\": \"...\"}, ...]\n",
|
181 |
+
" content = response.choices[0].message.content.strip()\n",
|
182 |
+
"\n",
|
183 |
+
"\n",
|
184 |
+
" return content\n",
|
185 |
+
"\n"
|
186 |
+
],
|
187 |
+
"metadata": {
|
188 |
+
"id": "wZhnhpHLSi7c"
|
189 |
+
},
|
190 |
+
"execution_count": 5,
|
191 |
+
"outputs": []
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"cell_type": "code",
|
195 |
+
"source": [
|
196 |
+
"faq_data = []\n",
|
197 |
+
"import re\n",
|
198 |
+
"for topic in FAQ_TOPICS:\n",
|
199 |
+
" questions=get_faq_ques_for_topic(topic)\n",
|
200 |
+
" raw_text = questions.strip()\n",
|
201 |
+
" cleaned = re.sub(r\"^```json|```$\", \"\", raw_text, flags=re.IGNORECASE).strip(\"`\\n \")\n",
|
202 |
+
"\n",
|
203 |
+
" # Now attempt to parse\n",
|
204 |
+
" question_data = json.loads(cleaned)\n",
|
205 |
+
" for key in question_data:\n",
|
206 |
+
" answer=get_faq_ans_for_topic(topic,key['question'])\n",
|
207 |
+
" faq_data.append({\n",
|
208 |
+
" \"topic\": topic,\n",
|
209 |
+
" \"question\": key['question'],\n",
|
210 |
+
" \"answer\": answer\n",
|
211 |
+
" })\n"
|
212 |
+
],
|
213 |
+
"metadata": {
|
214 |
+
"id": "bzTtCQijTlmq"
|
215 |
+
},
|
216 |
+
"execution_count": 6,
|
217 |
+
"outputs": []
|
218 |
+
},
|
219 |
+
{
|
220 |
+
"cell_type": "code",
|
221 |
+
"source": [
|
222 |
+
"import pandas as pd\n",
|
223 |
+
"pd.DataFrame(faq_data).to_csv(\"faq_data.csv\",index=False)"
|
224 |
+
],
|
225 |
+
"metadata": {
|
226 |
+
"id": "nKwJfsMEWstr"
|
227 |
+
},
|
228 |
+
"execution_count": 7,
|
229 |
+
"outputs": []
|
230 |
+
}
|
231 |
+
]
|
232 |
+
}
|