File size: 6,540 Bytes
a71b275 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 |
import streamlit as st
import pandas as pd
import plotly.express as px
from cryptography.fernet import Fernet
import time
import io
from transformers import pipeline
from streamlit_extras.stylable_container import stylable_container
import json
from io import StringIO
# sidebar
with st.sidebar:
with stylable_container(
key="test_button",
css_styles="""
button {
background-color: yellow;
border: 1px solid black;
padding: 5px;
color: black;
}
""",
):
st.button("DEMO APP")
expander = st.expander("**Important notes on the Google Sheet Table Question Answering (QA) App**")
expander.write('''
**Supported File Formats**
This app works with public URLs of Google Sheets. Google Sheets must not exceed 3,000 rows.
**How to Use**
Paste the public URL of your Google Sheet. Press the 'Fetch Data' button, then type your question into the text area provided and click the 'Retrieve your answer' button. Always press the 'Fetch Data' button before typing a question.
**Usage Limits**
You can ask up to 5 questions.
**Subscription Management**
This demo app offers a one-day subscription, expiring after 24 hours. If you are interested in building your own Table Question Answering (QA) Web App, we invite you to explore our NLP Web App Store on our website. You can select your desired features, place your order, and we will deliver your custom app in five business days. If you wish to delete your Account with us, please contact us at [email protected]
**Authorization**
For security purposes, your authorization access expires hourly. To restore access, click the "Request Authorization" button.
**Customization**
To change the app's background color to white or black, click the three-dot menu on the right-hand side of your app, go to Settings and then Choose app theme, colors and fonts.
**File Handling and Errors**
The app may display an error message if your file has errors.
For any errors or inquiries, please contact us at [email protected]
''')
if 'fernet_key' not in st.session_state:
st.session_state.fernet_key = Fernet.generate_key()
key = st.session_state.fernet_key
# function for generating and validating fernet key
def generate_fernet_token(key, data):
fernet = Fernet(key)
token = fernet.encrypt(data.encode())
return token
def validate_fernet_token(key, token, ttl_seconds):
fernet = Fernet(key)
try:
decrypted_data = fernet.decrypt(token, ttl=ttl_seconds).decode()
return decrypted_data, None
except Exception as e:
return None, f"Expired token: {e}"
if 'question_attempts' not in st.session_state:
st.session_state['question_attempts'] = 0
max_attempts = 5
def clear_text():
st.session_state["url"] = ""
# --- UI elements ---
st.subheader("Google Sheet Question Answering (QA)", divider = "green")
url = st.text_input("Enter Google Sheet URL:", key="url")
st.button("Clear URL", on_click=clear_text)
# --- Data fetching and processing ---
if st.button("Fetch Data"):
if url:
try:
spreadsheet_key = url.split('/d/')[1].split('/')[0]
csv_url = f'https://docs.google.com/spreadsheets/d/{spreadsheet_key}/export?format=csv'
df = pd.read_csv(csv_url, na_filter=False)
st.dataframe(df)
st.write("_number of rows_", df.shape[0])
st.write("_number of columns_", df.shape[1])
st.session_state.df = df # Store DataFrame in session state
if df.shape[0] > 3000: # Check row count inside try block
st.warning ("Google Sheets must not exceed 3,000 rows.")
st.stop()
except Exception as e:
st.error(f"Error fetching data: {e}")
else:
st.warning("Please enter a Google Sheet URL.")
st.divider()
# Authorization and question answering
if 'fernet_token' not in st.session_state:
if 'df' in st.session_state:
df = st.session_state.df
st.session_state.fernet_token = generate_fernet_token(key, df.to_json())
else:
st.stop()
decrypted_data_streamlit, error_streamlit = validate_fernet_token(key, st.session_state.fernet_token, ttl_seconds=3600)
if error_streamlit:
st.warning("Please press Request Authorization.")
if st.button("Request Authorization"):
if 'df' in st.session_state:
df = st.session_state.df
st.session_state.fernet_token = generate_fernet_token(key, df.to_json())
st.success("Authorization granted")
decrypted_data_streamlit, error_streamlit = validate_fernet_token(key, st.session_state.fernet_token, ttl_seconds=3600)
if error_streamlit:
st.error(f"Your authorization has expired: {error_streamlit}")
st.stop()
if error_streamlit:
st.error("Please paste the public URL of your Google Sheet.")
st.stop()
else:
try:
df = pd.read_json(decrypted_data_streamlit)
except Exception as e:
st.error(f"Error decoding data: {e}")
st.stop()
else:
st.error(f"Your authorization has expired: {error_streamlit}")
st.stop()
def clear_question():
st.session_state["question"] = ""
question = st.text_input("Type your question here:", key="question")
st.button("Clear question", on_click=clear_question)
if 'df' in st.session_state: # Check if DataFrame is in session state
df = st.session_state.df # Retrieve DataFrame from session state
if st.button("Retrieve your answer"):
if st.session_state.question_attempts < max_attempts:
try:
tqa = pipeline(task="table-question-answering", model="microsoft/tapex-large-finetuned-wtq")
answer = tqa(table=df, query=question)['answer']
st.write(f"Answer: {answer}")
st.session_state.question_attempts += 1
except Exception as e:
st.error(f"Error retrieving answer: {e}")
else:
st.error("Maximum question attempts reached.")
else:
st.warning("Please fetch data first.")
st.divider()
st.write(f"Number of times you asked a question: {st.session_state['question_attempts']}/{max_attempts}")
|