Spaces:
Sleeping
Sleeping
Commit
•
6e6b89c
1
Parent(s):
dc7e31c
Update app.py
Browse files
app.py
CHANGED
@@ -1,247 +1,90 @@
|
|
1 |
-
import
|
2 |
-
from typing import Iterator
|
3 |
-
|
4 |
import gradio as gr
|
5 |
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
DEFAULT_SYSTEM_PROMPT = "You are Mistral. You are AI-assistant, you are polite, give only truthful information and are based on the Mistral-7B model from Mistral AI. You can communicate in different languages equally well."
|
11 |
-
MAX_MAX_NEW_TOKENS = 4096
|
12 |
-
DEFAULT_MAX_NEW_TOKENS = 256
|
13 |
-
MAX_INPUT_TOKEN_LENGTH = 4000
|
14 |
-
|
15 |
-
DESCRIPTION = """
|
16 |
-
# [Mistral-7B](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1)
|
17 |
-
"""
|
18 |
-
|
19 |
-
def clear_and_save_textbox(message: str) -> tuple[str, str]:
|
20 |
-
return '', message
|
21 |
-
|
22 |
-
|
23 |
-
def display_input(message: str,
|
24 |
-
history: list[tuple[str, str]]) -> list[tuple[str, str]]:
|
25 |
-
history.append((message, ''))
|
26 |
-
return history
|
27 |
|
28 |
|
29 |
-
def
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
|
38 |
def generate(
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
top_p
|
45 |
-
|
46 |
-
|
47 |
-
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
first_response = next(generator)
|
54 |
-
yield history + [(message, first_response)]
|
55 |
-
except StopIteration:
|
56 |
-
yield history + [(message, '')]
|
57 |
-
for response in generator:
|
58 |
-
yield history + [(message, response)]
|
59 |
-
|
60 |
-
|
61 |
-
def process_example(message: str) -> tuple[str, list[tuple[str, str]]]:
|
62 |
-
generator = generate(message, [], DEFAULT_SYSTEM_PROMPT, 1024, 1, 0.95, 50)
|
63 |
-
for x in generator:
|
64 |
-
pass
|
65 |
-
return '', x
|
66 |
-
|
67 |
-
|
68 |
-
def check_input_token_length(message: str, chat_history: list[tuple[str, str]], system_prompt: str) -> None:
|
69 |
-
input_token_length = len(message) + len(chat_history)
|
70 |
-
if input_token_length > MAX_INPUT_TOKEN_LENGTH:
|
71 |
-
raise gr.Error(f'The accumulated input is too long ({input_token_length} > {MAX_INPUT_TOKEN_LENGTH}). Clear your chat history and try again.')
|
72 |
-
|
73 |
-
|
74 |
-
with gr.Blocks(css='style.css') as demo:
|
75 |
-
gr.Markdown(DESCRIPTION)
|
76 |
-
gr.DuplicateButton(value='Duplicate Space for private use',
|
77 |
-
elem_id='duplicate-button')
|
78 |
-
|
79 |
-
with gr.Group():
|
80 |
-
chatbot = gr.Chatbot(label='Playground')
|
81 |
-
with gr.Row():
|
82 |
-
textbox = gr.Textbox(
|
83 |
-
container=False,
|
84 |
-
show_label=False,
|
85 |
-
placeholder='Hi, Mistral!',
|
86 |
-
scale=10,
|
87 |
-
)
|
88 |
-
submit_button = gr.Button('Submit',
|
89 |
-
variant='primary',
|
90 |
-
scale=1,
|
91 |
-
min_width=0)
|
92 |
-
with gr.Row():
|
93 |
-
retry_button = gr.Button('🔄 Retry', variant='secondary')
|
94 |
-
undo_button = gr.Button('↩️ Undo', variant='secondary')
|
95 |
-
clear_button = gr.Button('🗑️ Clear', variant='secondary')
|
96 |
-
|
97 |
-
saved_input = gr.State()
|
98 |
-
|
99 |
-
with gr.Accordion(label='⚙️ Advanced options', open=False):
|
100 |
-
system_prompt = gr.Textbox(label='System prompt',
|
101 |
-
value=DEFAULT_SYSTEM_PROMPT,
|
102 |
-
lines=5,
|
103 |
-
interactive=False)
|
104 |
-
max_new_tokens = gr.Slider(
|
105 |
-
label='Max new tokens',
|
106 |
-
minimum=1,
|
107 |
-
maximum=MAX_MAX_NEW_TOKENS,
|
108 |
-
step=1,
|
109 |
-
value=DEFAULT_MAX_NEW_TOKENS,
|
110 |
-
)
|
111 |
-
temperature = gr.Slider(
|
112 |
-
label='Temperature',
|
113 |
-
minimum=0.1,
|
114 |
-
maximum=4.0,
|
115 |
-
step=0.1,
|
116 |
-
value=0.1,
|
117 |
-
)
|
118 |
-
top_p = gr.Slider(
|
119 |
-
label='Top-p (nucleus sampling)',
|
120 |
-
minimum=0.05,
|
121 |
-
maximum=1.0,
|
122 |
-
step=0.05,
|
123 |
-
value=0.9,
|
124 |
-
)
|
125 |
-
top_k = gr.Slider(
|
126 |
-
label='Top-k',
|
127 |
-
minimum=1,
|
128 |
-
maximum=1000,
|
129 |
-
step=1,
|
130 |
-
value=10,
|
131 |
-
)
|
132 |
-
|
133 |
-
|
134 |
-
|
135 |
-
textbox.submit(
|
136 |
-
fn=clear_and_save_textbox,
|
137 |
-
inputs=textbox,
|
138 |
-
outputs=[textbox, saved_input],
|
139 |
-
api_name=False,
|
140 |
-
queue=False,
|
141 |
-
).then(
|
142 |
-
fn=display_input,
|
143 |
-
inputs=[saved_input, chatbot],
|
144 |
-
outputs=chatbot,
|
145 |
-
api_name=False,
|
146 |
-
queue=False,
|
147 |
-
).then(
|
148 |
-
fn=check_input_token_length,
|
149 |
-
inputs=[saved_input, chatbot, system_prompt],
|
150 |
-
api_name=False,
|
151 |
-
queue=False,
|
152 |
-
).success(
|
153 |
-
fn=generate,
|
154 |
-
inputs=[
|
155 |
-
saved_input,
|
156 |
-
chatbot,
|
157 |
-
system_prompt,
|
158 |
-
max_new_tokens,
|
159 |
-
temperature,
|
160 |
-
top_p,
|
161 |
-
top_k,
|
162 |
-
],
|
163 |
-
outputs=chatbot,
|
164 |
-
api_name=False,
|
165 |
-
)
|
166 |
-
|
167 |
-
button_event_preprocess = submit_button.click(
|
168 |
-
fn=clear_and_save_textbox,
|
169 |
-
inputs=textbox,
|
170 |
-
outputs=[textbox, saved_input],
|
171 |
-
api_name=False,
|
172 |
-
queue=False,
|
173 |
-
).then(
|
174 |
-
fn=display_input,
|
175 |
-
inputs=[saved_input, chatbot],
|
176 |
-
outputs=chatbot,
|
177 |
-
api_name=False,
|
178 |
-
queue=False,
|
179 |
-
).then(
|
180 |
-
fn=check_input_token_length,
|
181 |
-
inputs=[saved_input, chatbot, system_prompt],
|
182 |
-
api_name=False,
|
183 |
-
queue=False,
|
184 |
-
).success(
|
185 |
-
fn=generate,
|
186 |
-
inputs=[
|
187 |
-
saved_input,
|
188 |
-
chatbot,
|
189 |
-
system_prompt,
|
190 |
-
max_new_tokens,
|
191 |
-
temperature,
|
192 |
-
top_p,
|
193 |
-
top_k,
|
194 |
-
],
|
195 |
-
outputs=chatbot,
|
196 |
-
api_name=False,
|
197 |
)
|
198 |
|
199 |
-
|
200 |
-
|
201 |
-
|
202 |
-
|
203 |
-
|
204 |
-
|
205 |
-
|
206 |
-
|
207 |
-
|
208 |
-
|
209 |
-
|
210 |
-
|
211 |
-
|
212 |
-
|
213 |
-
|
214 |
-
|
215 |
-
|
216 |
-
|
217 |
-
|
218 |
-
|
219 |
-
|
220 |
-
|
221 |
-
|
222 |
-
|
223 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
224 |
)
|
|
|
225 |
|
226 |
-
undo_button.click(
|
227 |
-
fn=delete_prev_fn,
|
228 |
-
inputs=chatbot,
|
229 |
-
outputs=[chatbot, saved_input],
|
230 |
-
api_name=False,
|
231 |
-
queue=False,
|
232 |
-
).then(
|
233 |
-
fn=lambda x: x,
|
234 |
-
inputs=[saved_input],
|
235 |
-
outputs=textbox,
|
236 |
-
api_name=False,
|
237 |
-
queue=False,
|
238 |
-
)
|
239 |
-
|
240 |
-
clear_button.click(
|
241 |
-
fn=lambda: ([], ''),
|
242 |
-
outputs=[chatbot, saved_input],
|
243 |
-
queue=False,
|
244 |
-
api_name=False,
|
245 |
-
)
|
246 |
|
247 |
-
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
from huggingface_hub import InferenceClient
|
|
|
|
|
2 |
import gradio as gr
|
3 |
|
4 |
+
client = InferenceClient(
|
5 |
+
"mistralai/Mistral-7B-Instruct-v0.2"
|
6 |
+
)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
|
8 |
|
9 |
+
def format_prompt(message, history):
|
10 |
+
prompt = "<s>"
|
11 |
+
for user_prompt, bot_response in history:
|
12 |
+
prompt += f"[INST] {user_prompt} [/INST]"
|
13 |
+
prompt += f" {bot_response}</s> "
|
14 |
+
prompt += f"[INST] {message} [/INST]"
|
15 |
+
return prompt
|
|
|
16 |
|
17 |
def generate(
|
18 |
+
prompt, history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
|
19 |
+
):
|
20 |
+
temperature = float(temperature)
|
21 |
+
if temperature < 1e-2:
|
22 |
+
temperature = 1e-2
|
23 |
+
top_p = float(top_p)
|
24 |
+
|
25 |
+
generate_kwargs = dict(
|
26 |
+
temperature=temperature,
|
27 |
+
max_new_tokens=max_new_tokens,
|
28 |
+
top_p=top_p,
|
29 |
+
repetition_penalty=repetition_penalty,
|
30 |
+
do_sample=True,
|
31 |
+
seed=42,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
)
|
33 |
|
34 |
+
formatted_prompt = format_prompt(prompt, history)
|
35 |
+
|
36 |
+
stream = client.text_generation(formatted_prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
|
37 |
+
output = ""
|
38 |
+
|
39 |
+
for response in stream:
|
40 |
+
output += response.token.text
|
41 |
+
yield output
|
42 |
+
return output
|
43 |
+
|
44 |
+
|
45 |
+
additional_inputs=[
|
46 |
+
gr.Slider(
|
47 |
+
label="Temperature",
|
48 |
+
value=0.9,
|
49 |
+
minimum=0.0,
|
50 |
+
maximum=1.0,
|
51 |
+
step=0.05,
|
52 |
+
interactive=True,
|
53 |
+
info="Higher values produce more diverse outputs",
|
54 |
+
),
|
55 |
+
gr.Slider(
|
56 |
+
label="Max new tokens",
|
57 |
+
value=256,
|
58 |
+
minimum=0,
|
59 |
+
maximum=1048,
|
60 |
+
step=64,
|
61 |
+
interactive=True,
|
62 |
+
info="The maximum numbers of new tokens",
|
63 |
+
),
|
64 |
+
gr.Slider(
|
65 |
+
label="Top-p (nucleus sampling)",
|
66 |
+
value=0.90,
|
67 |
+
minimum=0.0,
|
68 |
+
maximum=1,
|
69 |
+
step=0.05,
|
70 |
+
interactive=True,
|
71 |
+
info="Higher values sample more low-probability tokens",
|
72 |
+
),
|
73 |
+
gr.Slider(
|
74 |
+
label="Repetition penalty",
|
75 |
+
value=1.2,
|
76 |
+
minimum=1.0,
|
77 |
+
maximum=2.0,
|
78 |
+
step=0.05,
|
79 |
+
interactive=True,
|
80 |
+
info="Penalize repeated tokens",
|
81 |
)
|
82 |
+
]
|
83 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
84 |
|
85 |
+
gr.ChatInterface(
|
86 |
+
fn=generate,
|
87 |
+
chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel"),
|
88 |
+
additional_inputs=additional_inputs,
|
89 |
+
title="""Mistral 7B ```v0.2```"""
|
90 |
+
).launch(show_api=False)
|