Spaces:
Runtime error
Runtime error
File size: 5,998 Bytes
b45f299 1a83b46 b45f299 e6cbc32 b45f299 4fb337d b45f299 029c3b0 b45f299 cf34b99 b45f299 8971fda 429e239 b45f299 429e239 b45f299 8971fda 4fb337d 8971fda b45f299 8971fda b45f299 8971fda b45f299 4fb337d b45f299 4fb337d 88c05b4 8971fda 88c05b4 656338b aac2baa 88c05b4 b45f299 517830a b168ade c8f3309 6130bd5 b45f299 6130bd5 b45f299 4fb337d b45f299 4fb337d 83166a1 4fb337d de27691 4fb337d b45f299 e6cbc32 7eddd34 b45f299 |
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 os
import requests
import json
import gradio as gr
from transformers import AutoTokenizer
DESCRIPTION = """
"""
LICENSE = """
"""
DEFAULT_SYSTEM_PROMPT = "You are a helpful AI assistant built by MediaTek Research. The user you are helping speaks Traditional Chinese and comes from Taiwan."
API_URL = os.environ.get("API_URL")
TOKEN = os.environ.get("TOKEN")
HEADERS = {
"accept": "application/json",
"Authorization": f"Bearer {TOKEN}",
"Content-Type": "application/json",
}
MODEL_NAME="breeze-7b-instruct-v01"
PRESENCE_PENALTY=0
FREQUENCY_PENALTY=0
model_name = "MediaTek-Research/Breeze-7B-Instruct-v0.1"
tokenizer = AutoTokenizer.from_pretrained(model_name)
with gr.Blocks() as demo:
gr.Markdown(DESCRIPTION)
chatbot = gr.Chatbot()
with gr.Row():
msg = gr.Textbox(
container=False,
show_label=False,
placeholder='Type a message...',
scale=10,
)
submit_button = gr.Button('Submit',
variant='primary',
scale=1,
min_width=0)
with gr.Row():
retry_button = gr.Button('🔄 Retry', variant='secondary')
undo_button = gr.Button('↩️ Undo', variant='secondary')
clear = gr.Button('🗑️ Clear', variant='secondary')
saved_input = gr.State()
with gr.Accordion(label='Advanced options', open=False):
system_prompt = gr.Textbox(label='System prompt',
value=DEFAULT_SYSTEM_PROMPT,
lines=6)
max_new_tokens = gr.Slider(
label='Max new tokens',
minimum=32,
maximum=1024,
step=1,
value=512,
)
temperature = gr.Slider(
label='Temperature',
minimum=0.01,
maximum=0.5,
step=0.01,
value=0.01,
)
top_p = gr.Slider(
label='Top-p (nucleus sampling)',
minimum=0.01,
maximum=1.0,
step=0.01,
value=0.01,
)
def user(user_message, history):
return "", history + [[user_message, None]]
def bot(history, max_new_tokens, temperature, top_p, system_prompt):
chat_data = []
system_prompt = system_prompt.strip()
if system_prompt:
chat_data.append({"role": "system", "content": system_prompt})
for user_msg, assistant_msg in history:
if user_msg is not None:
chat_data.append({"role": "user", "content": user_msg})
if assistant_msg is not None:
chat_data.append({"role": "assistant", "content": assistant_msg})
message = tokenizer.apply_chat_template(chat_data, tokenize=False)
message = message[3:] # remove SOT token
data = {
"model": MODEL_NAME,
"prompt": str(message),
"temperature": float(temperature) + 0.01,
"n": 1,
"max_tokens": int(max_new_tokens),
"stop": "",
"top_p": float(top_p),
"logprobs": 0,
"echo": False,
"presence_penalty": PRESENCE_PENALTY,
"frequency_penalty": FREQUENCY_PENALTY,
"stream": True,
}
with requests.post(API_URL, headers=HEADERS, data=json.dumps(data), stream=True) as r:
for response in r.iter_lines():
print(response)
if len(response) > 0:
text = response.decode()
if text != "data: [DONE]":
if text.startswith("data: "):
text = text[5:]
delta = json.loads(text)["choices"][0]["text"]
if history[-1][1] is None:
history[-1][1] = delta
else:
history[-1][1] += delta
yield history
msg.submit(user, [msg, chatbot], [msg, chatbot], queue=False).then(
fn=bot,
inputs=[
chatbot,
max_new_tokens,
temperature,
top_p,
system_prompt,
],
outputs=chatbot
)
submit_button.click(
user, [msg, chatbot], [msg, chatbot], queue=False
).then(
fn=bot,
inputs=[
chatbot,
max_new_tokens,
temperature,
top_p,
system_prompt,
],
outputs=chatbot
)
def delete_prev_fn(
history: list[tuple[str, str]]) -> tuple[list[tuple[str, str]], str]:
try:
message, _ = history.pop()
except IndexError:
message = ''
return history, message or ''
def display_input(message: str,
history: list[tuple[str, str]]) -> list[tuple[str, str]]:
history.append((message, ''))
return history
retry_button.click(
fn=delete_prev_fn,
inputs=chatbot,
outputs=[chatbot, saved_input],
api_name=False,
queue=False,
).then(
fn=display_input,
inputs=[saved_input, chatbot],
outputs=chatbot,
api_name=False,
queue=False,
).then(
fn=bot,
inputs=[
chatbot,
max_new_tokens,
temperature,
top_p,
system_prompt,
],
outputs=chatbot,
)
undo_button.click(
fn=delete_prev_fn,
inputs=chatbot,
outputs=[chatbot, saved_input],
api_name=False,
queue=False,
).then(
fn=lambda x: x,
inputs=[saved_input],
outputs=msg,
api_name=False,
queue=False,
)
clear.click(lambda: None, None, chatbot, queue=False)
gr.Markdown(LICENSE)
# demo.queue(concurrency_count=4, max_size=128)
demo.launch()
|