forked from a16z-infra/llama2-chatbot
-
Notifications
You must be signed in to change notification settings - Fork 0
/
llama2_chatbot.py
176 lines (150 loc) · 7.02 KB
/
llama2_chatbot.py
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
"""
LLaMA 2 Chatbot app
======================
This is an Streamlit chatbot app with LLaMA2 that includes session chat history and option to select multiple LLM
API enpoints on Replicate. Each model (7B & 13B) runs on Replicate on one A100 (40Gb). The weights have been tensorized.
Author: Marco Mascorro (@mascobot.com)
Created: July 2023
Version: 0.9.0 (Experimental)
Status: Development
Python version: 3.9.15
a16z-infra
"""
#External libraries:
import streamlit as st
import replicate
from dotenv import load_dotenv
load_dotenv()
import os
from utils import debounce_replicate_run
# feel free to replace with your own logo
logo1 = 'https://storage.googleapis.com/llama2_release/a16z_logo.png'
logo2 = 'https://storage.googleapis.com/llama2_release/replicate_logo_white.png'
###Initial UI configuration:###
st.set_page_config(page_title="LLaMA2 Chatbot by a16z-infra", page_icon=logo1, layout="wide")
# reduce font sizes for input text boxes
custom_css = """
<style>
.stTextArea textarea {font-size: 13px;}
div[data-baseweb="select"] > div {font-size: 13px !important;}
</style>
"""
st.markdown(custom_css, unsafe_allow_html=True)
st.sidebar.header("LLaMA2 Chatbot")#Left sidebar menu
#Set config for a cleaner menu, footer & background:
hide_streamlit_style = """
<style>
#MainMenu {visibility: hidden;}
footer {visibility: hidden;}
</style>
"""
st.markdown(hide_streamlit_style, unsafe_allow_html=True)
###Global variables:###
REPLICATE_API_TOKEN = os.environ.get('REPLICATE_API_TOKEN', default='')
#Your your (Replicate) models' endpoints:
REPLICATE_MODEL_ENDPOINT7B = os.environ.get('REPLICATE_MODEL_ENDPOINT7B', default='')
REPLICATE_MODEL_ENDPOINT13B = os.environ.get('REPLICATE_MODEL_ENDPOINT13B', default='')
PRE_PROMPT = "You are a helpful assistant. You do not respond as 'User' or pretend to be 'User'. You only respond once as Assistant."
#container for the chat history
response_container = st.container()
#container for the user's text input
container = st.container()
#Set up/Initialize Session State variables:
if 'chat_dialogue' not in st.session_state:
st.session_state['chat_dialogue'] = []
if 'llm' not in st.session_state:
st.session_state['llm'] = REPLICATE_MODEL_ENDPOINT13B
if 'temperature' not in st.session_state:
st.session_state['temperature'] = 0.1
if 'top_p' not in st.session_state:
st.session_state['top_p'] = 0.9
if 'max_seq_len' not in st.session_state:
st.session_state['max_seq_len'] = 512
if 'pre_prompt' not in st.session_state:
st.session_state['pre_prompt'] = PRE_PROMPT
if 'string_dialogue' not in st.session_state:
st.session_state['string_dialogue'] = ''
#Dropdown menu to select the model edpoint:
selected_option = st.sidebar.selectbox('Choose a LLaMA2 model:', ['LLaMA2-13B', 'LLaMA2-7B'], key='model')
if selected_option == 'LLaMA2-7B':
st.session_state['llm'] = REPLICATE_MODEL_ENDPOINT7B
else:
st.session_state['llm'] = REPLICATE_MODEL_ENDPOINT13B
#Model hyper parameters:
st.session_state['temperature'] = st.sidebar.slider('Temperature:', min_value=0.01, max_value=5.0, value=0.1, step=0.01)
st.session_state['top_p'] = st.sidebar.slider('Top P:', min_value=0.01, max_value=1.0, value=0.9, step=0.01)
st.session_state['max_seq_len'] = st.sidebar.slider('Max Sequence Length:', min_value=64, max_value=4096, value=2048, step=8)
NEW_P = st.sidebar.text_area('Prompt before the chat starts. Edit here if desired:', PRE_PROMPT, height=60)
if NEW_P != PRE_PROMPT and NEW_P != "" and NEW_P != None:
st.session_state['pre_prompt'] = NEW_P + "\n\n"
else:
st.session_state['pre_prompt'] = PRE_PROMPT
# Add the "Clear Chat History" button to the sidebar
clear_chat_history_button = st.sidebar.button("Clear Chat History")
# Check if the button is clicked
if clear_chat_history_button:
# Reset the chat history stored in the session state
st.session_state['chat_dialogue'] = []
# add links to relevant resources for users to select
text1 = 'Chatbot Demo Code'
text2 = 'Model on Replicate'
text3 = 'LLaMa2 Cog Template'
logo1_link = "https://github.com/a16z-infra/llama2-chatbot"
logo2_link = "https://replicate.com/a16z-infra/llama13b-v2-chat"
text3_link = "https://github.com/a16z-infra/cog-llama-template"
st.sidebar.markdown(f"""
<div class='resources-section'>
<h3>Resources:</h3>
<div style="display: flex; justify-content: space-between;">
<div style="display: flex; flex-direction: column; padding-left: 15px;">
<div style="align-self: flex-start; padding-bottom: 5px;"> <!-- Change to flex-start here -->
<a href="{logo1_link}">
<img src="{logo1}" alt="Logo 1" style="width: 30px;"/>
</a>
</div>
<div style="align-self: flex-start;">
<p style="font-size:11px; margin-bottom: -5px;"><a href="{logo1_link}">{text1}</a></p>
<p style="font-size:11px;"><a href="{text3_link}">{text3}</a></p> <!-- second line of text -->
</div>
</div>
<div style="display: flex; flex-direction: column; padding-right: 25px;">
<div style="align-self: flex-start; padding-bottom: 5px;">
<a href="{logo2_link}">
<img src="{logo2}" alt="Logo 2" style="width: 120px;"/>
</a>
</div>
<div style="align-self: flex-start;">
<p style="font-size:11px;"><a href="{logo2_link}">{text2}</a></p>
</div>
</div>
</div>
</div>
""", unsafe_allow_html=True)
# Display chat messages from history on app rerun
for message in st.session_state.chat_dialogue:
with st.chat_message(message["role"]):
st.markdown(message["content"])
# Accept user input
if prompt := st.chat_input("Type your question here to talk to LLaMA2"):
# Add user message to chat history
st.session_state.chat_dialogue.append({"role": "user", "content": prompt})
# Display user message in chat message container
with st.chat_message("user"):
st.markdown(prompt)
with st.chat_message("assistant"):
message_placeholder = st.empty()
full_response = ""
string_dialogue = st.session_state['pre_prompt']
for dict_message in st.session_state.chat_dialogue:
if dict_message["role"] == "user":
string_dialogue = string_dialogue + "User: " + dict_message["content"] + "\n\n"
else:
string_dialogue = string_dialogue + "Assistant: " + dict_message["content"] + "\n\n"
print (string_dialogue)
output = debounce_replicate_run(st.session_state['llm'], string_dialogue + "Assistant: ", st.session_state['max_seq_len'], st.session_state['temperature'], st.session_state['top_p'], REPLICATE_API_TOKEN)
for item in output:
full_response += item
message_placeholder.markdown(full_response + "▌")
message_placeholder.markdown(full_response)
# Add assistant response to chat history
st.session_state.chat_dialogue.append({"role": "assistant", "content": full_response})