added IT models for chat
Browse files
app.py
CHANGED
@@ -1,19 +1,17 @@
|
|
1 |
-
from threading import Thread
|
2 |
-
|
3 |
import gradio as gr
|
4 |
-
from
|
5 |
-
import spaces
|
6 |
from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
|
7 |
import torch
|
8 |
-
|
9 |
from open_lm.hf import *
|
10 |
from open_lm.precision import get_autocast
|
11 |
-
|
12 |
|
13 |
# Define model options
|
14 |
MODEL_OPTIONS = {
|
15 |
"TRI DCLM-1B": "TRI-ML/DCLM-1B",
|
16 |
-
"Apple DCLM-Baseline-7B": "apple/DCLM-Baseline-7B"
|
|
|
|
|
17 |
}
|
18 |
|
19 |
# Global variables for model and tokenizer
|
@@ -29,13 +27,13 @@ def load_model(model_name):
|
|
29 |
return f"Loaded model: {model_name}"
|
30 |
|
31 |
@spaces.GPU
|
32 |
-
def
|
33 |
prompt, model_choice, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
|
34 |
):
|
35 |
global current_model, current_tokenizer
|
36 |
|
37 |
if current_model is None or current_tokenizer is None:
|
38 |
-
return "Please
|
39 |
|
40 |
temperature = float(temperature)
|
41 |
if temperature < 1e-2:
|
@@ -63,7 +61,6 @@ def generate(
|
|
63 |
thread = Thread(target=current_model.generate, kwargs=generate_kwargs)
|
64 |
thread.start()
|
65 |
|
66 |
-
# Write the prompt in blue
|
67 |
output = "<span style='color: blue;'>" + prompt + "</span>"
|
68 |
for new_text in streamer:
|
69 |
if isinstance(new_text, torch.Tensor):
|
@@ -77,80 +74,159 @@ def generate(
|
|
77 |
thread.join()
|
78 |
return output
|
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 |
with gr.Blocks() as demo:
|
120 |
gr.Markdown(
|
121 |
"""
|
122 |
-
# DCLM
|
123 |
-
This demo allows you to generate text using
|
124 |
-
|
|
|
|
|
|
|
125 |
|
126 |
-
|
127 |
-
|
128 |
"""
|
129 |
)
|
130 |
|
131 |
-
|
132 |
with gr.Row():
|
133 |
model_dropdown = gr.Dropdown(choices=list(MODEL_OPTIONS.keys()), label="Select Model")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
134 |
|
135 |
-
model_dropdown.
|
136 |
-
|
137 |
inputs=[model_dropdown],
|
138 |
-
outputs=[
|
139 |
)
|
140 |
|
141 |
-
text_input = gr.Textbox(lines=3, label="Input Text")
|
142 |
-
text_output = gr.Markdown(label="Generated Text")
|
143 |
-
|
144 |
-
generate_button = gr.Button("Generate")
|
145 |
-
|
146 |
generate_button.click(
|
147 |
-
|
148 |
inputs=[text_input, model_dropdown, *additional_inputs],
|
149 |
outputs=[text_output]
|
150 |
)
|
151 |
-
with Accordion(label="Advanced Options", open=False):
|
152 |
-
for input_component in additional_inputs:
|
153 |
-
if not input_component.is_rendered:
|
154 |
-
input_component.render()
|
155 |
|
156 |
-
|
|
|
|
|
|
|
|
|
|
|
|
1 |
import gradio as gr
|
2 |
+
from threading import Thread
|
|
|
3 |
from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
|
4 |
import torch
|
|
|
5 |
from open_lm.hf import *
|
6 |
from open_lm.precision import get_autocast
|
7 |
+
import spaces
|
8 |
|
9 |
# Define model options
|
10 |
MODEL_OPTIONS = {
|
11 |
"TRI DCLM-1B": "TRI-ML/DCLM-1B",
|
12 |
+
"Apple DCLM-Baseline-7B": "apple/DCLM-Baseline-7B",
|
13 |
+
"[IT] TRI DCLM-1B": "TRI-ML/DCLM-1B-IT",
|
14 |
+
"[IT] Apple DCLM-Baseline-7B": "mlfoundations/dclm-7b-it",
|
15 |
}
|
16 |
|
17 |
# Global variables for model and tokenizer
|
|
|
27 |
return f"Loaded model: {model_name}"
|
28 |
|
29 |
@spaces.GPU
|
30 |
+
def generate_completion(
|
31 |
prompt, model_choice, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
|
32 |
):
|
33 |
global current_model, current_tokenizer
|
34 |
|
35 |
if current_model is None or current_tokenizer is None:
|
36 |
+
return "Please select a model first."
|
37 |
|
38 |
temperature = float(temperature)
|
39 |
if temperature < 1e-2:
|
|
|
61 |
thread = Thread(target=current_model.generate, kwargs=generate_kwargs)
|
62 |
thread.start()
|
63 |
|
|
|
64 |
output = "<span style='color: blue;'>" + prompt + "</span>"
|
65 |
for new_text in streamer:
|
66 |
if isinstance(new_text, torch.Tensor):
|
|
|
74 |
thread.join()
|
75 |
return output
|
76 |
|
77 |
+
def format_prompt(message, history):
|
78 |
+
prompt = ""
|
79 |
+
for user_prompt, bot_response in history:
|
80 |
+
prompt += f"User: {user_prompt}\nAssistant: {bot_response}\n"
|
81 |
+
prompt += f"User: {message}\nAssistant:"
|
82 |
+
return prompt
|
83 |
+
|
84 |
+
@spaces.GPU
|
85 |
+
def generate_chat(
|
86 |
+
message, chat_history, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
|
87 |
+
):
|
88 |
+
global current_model, current_tokenizer
|
89 |
+
|
90 |
+
if current_model is None or current_tokenizer is None:
|
91 |
+
yield chat_history + [("Error", "Please select a model first.")]
|
92 |
+
return
|
93 |
+
|
94 |
+
temperature = float(temperature)
|
95 |
+
if temperature < 1e-2:
|
96 |
+
temperature = 1e-2
|
97 |
+
top_p = float(top_p)
|
98 |
+
|
99 |
+
formatted_prompt = format_prompt(message, chat_history)
|
100 |
+
inputs = current_tokenizer(formatted_prompt, return_tensors="pt").to(current_model.device)
|
101 |
+
|
102 |
+
generate_kwargs = dict(
|
103 |
+
**inputs,
|
104 |
+
max_new_tokens=max_new_tokens,
|
105 |
+
temperature=temperature,
|
106 |
+
top_p=top_p,
|
107 |
+
repetition_penalty=repetition_penalty,
|
108 |
+
do_sample=True,
|
109 |
+
pad_token_id=current_tokenizer.eos_token_id
|
110 |
+
)
|
111 |
+
|
112 |
+
streamer = TextIteratorStreamer(current_tokenizer, skip_prompt=True, skip_special_tokens=False)
|
113 |
+
streamer.stop_signal = current_tokenizer.decode(current_tokenizer.eos_token_id)
|
114 |
+
generate_kwargs["streamer"] = streamer
|
115 |
+
|
116 |
+
thread = Thread(target=current_model.generate, kwargs=generate_kwargs)
|
117 |
+
thread.start()
|
118 |
+
|
119 |
+
new_history = chat_history + [(message, "")]
|
120 |
+
for new_text in streamer:
|
121 |
+
if isinstance(new_text, torch.Tensor):
|
122 |
+
new_text = current_tokenizer.decode(new_text)
|
123 |
+
if streamer.stop_signal in new_text:
|
124 |
+
new_text = new_text.split(streamer.stop_signal)[0]
|
125 |
+
new_history[-1] = (message, new_history[-1][1] + new_text)
|
126 |
+
break
|
127 |
+
new_history[-1] = (message, new_history[-1][1] + new_text)
|
128 |
+
yield new_history
|
129 |
+
|
130 |
+
thread.join()
|
131 |
+
|
132 |
+
additional_inputs = [
|
133 |
+
gr.Slider(
|
134 |
+
label="Temperature",
|
135 |
+
value=0.9,
|
136 |
+
minimum=0.0,
|
137 |
+
maximum=1.0,
|
138 |
+
step=0.05,
|
139 |
+
interactive=True,
|
140 |
+
info="Higher values produce more diverse outputs",
|
141 |
+
),
|
142 |
+
gr.Slider(
|
143 |
+
label="Max new tokens",
|
144 |
+
value=256,
|
145 |
+
minimum=0,
|
146 |
+
maximum=1048,
|
147 |
+
step=64,
|
148 |
+
interactive=True,
|
149 |
+
info="The maximum numbers of new tokens",
|
150 |
+
),
|
151 |
+
gr.Slider(
|
152 |
+
label="Top-p (nucleus sampling)",
|
153 |
+
value=0.90,
|
154 |
+
minimum=0.0,
|
155 |
+
maximum=1,
|
156 |
+
step=0.05,
|
157 |
+
interactive=True,
|
158 |
+
info="Higher values sample more low-probability tokens",
|
159 |
+
),
|
160 |
+
gr.Slider(
|
161 |
+
label="Repetition penalty",
|
162 |
+
value=1.2,
|
163 |
+
minimum=1.0,
|
164 |
+
maximum=2.0,
|
165 |
+
step=0.05,
|
166 |
+
interactive=True,
|
167 |
+
info="Penalize repeated tokens",
|
168 |
+
)
|
169 |
+
]
|
170 |
|
171 |
with gr.Blocks() as demo:
|
172 |
gr.Markdown(
|
173 |
"""
|
174 |
+
# DCLM Demo
|
175 |
+
This demo allows you to generate text using DCLM models in two modes:
|
176 |
+
1. Text Completion:
|
177 |
+
For non-Instruction-Tuned models, it generates the continuation of the input text.
|
178 |
+
2. Chatbot:
|
179 |
+
For Instruction-Tuned [IT] models, it generates responses to user messages as a chatbot.
|
180 |
|
181 |
+
Select a model from the dropdown to start, it might take a few seconds to load.
|
182 |
+
The interface will automatically switch between Text Completion and Chatbot modes based on the selected model.
|
183 |
"""
|
184 |
)
|
185 |
|
|
|
186 |
with gr.Row():
|
187 |
model_dropdown = gr.Dropdown(choices=list(MODEL_OPTIONS.keys()), label="Select Model")
|
188 |
+
model_status = gr.Textbox(label="Model Status")
|
189 |
+
|
190 |
+
# Text Completion interface
|
191 |
+
with gr.Row(visible=False) as completion_interface:
|
192 |
+
with gr.Column():
|
193 |
+
text_input = gr.Textbox(lines=3, label="Input Text")
|
194 |
+
text_output = gr.Markdown(label="Generated Text")
|
195 |
+
generate_button = gr.Button("Generate")
|
196 |
+
|
197 |
+
# Chatbot interface
|
198 |
+
with gr.Row(visible=False) as chat_interface:
|
199 |
+
with gr.Column():
|
200 |
+
chatbot = gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel")
|
201 |
+
msg = gr.Textbox(label="Message")
|
202 |
+
clear = gr.Button("Clear")
|
203 |
+
|
204 |
+
with gr.Accordion("Advanced Options", open=False):
|
205 |
+
for input_component in additional_inputs:
|
206 |
+
input_component.render()
|
207 |
+
|
208 |
+
def switch_interface(model_name):
|
209 |
+
is_it_model = model_name.startswith("[IT]")
|
210 |
+
status = load_model(model_name)
|
211 |
+
return (
|
212 |
+
gr.Row(visible=not is_it_model), # completion_interface
|
213 |
+
gr.Row(visible=is_it_model), # chat_interface
|
214 |
+
status # model_status
|
215 |
+
)
|
216 |
|
217 |
+
model_dropdown.change(
|
218 |
+
switch_interface,
|
219 |
inputs=[model_dropdown],
|
220 |
+
outputs=[completion_interface, chat_interface, model_status]
|
221 |
)
|
222 |
|
|
|
|
|
|
|
|
|
|
|
223 |
generate_button.click(
|
224 |
+
generate_completion,
|
225 |
inputs=[text_input, model_dropdown, *additional_inputs],
|
226 |
outputs=[text_output]
|
227 |
)
|
|
|
|
|
|
|
|
|
228 |
|
229 |
+
msg.submit(generate_chat, [msg, chatbot, *additional_inputs], chatbot)
|
230 |
+
clear.click(lambda: None, None, chatbot, queue=False)
|
231 |
+
|
232 |
+
demo.queue().launch()
|