Spaces:
Running
Running
Update app.py
Browse files
app.py
CHANGED
@@ -1,7 +1,8 @@
|
|
1 |
import os
|
2 |
import threading
|
3 |
from collections import defaultdict
|
4 |
-
|
|
|
5 |
import gradio as gr
|
6 |
from transformers import (
|
7 |
AutoModelForCausalLM,
|
@@ -9,17 +10,14 @@ from transformers import (
|
|
9 |
TextIteratorStreamer,
|
10 |
)
|
11 |
|
12 |
-
# Define model paths
|
13 |
model_name_to_path = {
|
14 |
"LeCarnet-3M": "MaxLSB/LeCarnet-3M",
|
15 |
"LeCarnet-8M": "MaxLSB/LeCarnet-8M",
|
16 |
"LeCarnet-21M": "MaxLSB/LeCarnet-21M",
|
17 |
}
|
18 |
|
19 |
-
# Load Hugging Face token
|
20 |
hf_token = os.environ["HUGGINGFACEHUB_API_TOKEN"]
|
21 |
|
22 |
-
# Preload models and tokenizers
|
23 |
loaded_models = defaultdict(dict)
|
24 |
|
25 |
for name, path in model_name_to_path.items():
|
@@ -27,40 +25,24 @@ for name, path in model_name_to_path.items():
|
|
27 |
loaded_models[name]["model"] = AutoModelForCausalLM.from_pretrained(path, token=hf_token)
|
28 |
loaded_models[name]["model"].eval()
|
29 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
30 |
def respond(message, history, model_name, max_tokens, temperature, top_p):
|
31 |
-
"""
|
32 |
-
Generate a response from the selected model, streaming the output and updating chat history.
|
33 |
-
|
34 |
-
Args:
|
35 |
-
message (str): User's input message.
|
36 |
-
history (list): Current chat history as list of (user_msg, bot_msg) tuples.
|
37 |
-
model_name (str): Selected model name.
|
38 |
-
max_tokens (int): Maximum number of tokens to generate.
|
39 |
-
temperature (float): Sampling temperature.
|
40 |
-
top_p (float): Top-p sampling parameter.
|
41 |
-
|
42 |
-
Yields:
|
43 |
-
list: Updated chat history with the user's message and streaming bot response.
|
44 |
-
"""
|
45 |
-
# Append user's message to history with an empty bot response
|
46 |
history = history + [(message, "")]
|
47 |
-
yield history
|
48 |
-
|
49 |
-
# Select tokenizer and model
|
50 |
tokenizer = loaded_models[model_name]["tokenizer"]
|
51 |
model = loaded_models[model_name]["model"]
|
52 |
-
|
53 |
-
# Tokenize input
|
54 |
inputs = tokenizer(message, return_tensors="pt")
|
55 |
-
|
56 |
-
# Set up streaming
|
57 |
streamer = TextIteratorStreamer(
|
58 |
tokenizer,
|
59 |
skip_prompt=False,
|
60 |
skip_special_tokens=True,
|
61 |
)
|
62 |
-
|
63 |
-
# Configure generation parameters
|
64 |
generate_kwargs = dict(
|
65 |
**inputs,
|
66 |
streamer=streamer,
|
@@ -70,115 +52,61 @@ def respond(message, history, model_name, max_tokens, temperature, top_p):
|
|
70 |
top_p=top_p,
|
71 |
eos_token_id=tokenizer.eos_token_id,
|
72 |
)
|
73 |
-
|
74 |
-
# Start generation in a background thread
|
75 |
thread = threading.Thread(target=model.generate, kwargs=generate_kwargs)
|
76 |
thread.start()
|
77 |
-
|
78 |
-
# Stream the response with model name prefix
|
79 |
-
accumulated = f"**{model_name}:** "
|
80 |
for new_text in streamer:
|
81 |
accumulated += new_text
|
82 |
history[-1] = (message, accumulated)
|
83 |
yield history
|
84 |
|
85 |
def submit(message, history, model_name, max_tokens, temperature, top_p):
|
86 |
-
"""
|
87 |
-
Handle form submission by calling respond and clearing the input box.
|
88 |
-
|
89 |
-
Args:
|
90 |
-
message (str): User's input message.
|
91 |
-
history (list): Current chat history.
|
92 |
-
model_name (str): Selected model name.
|
93 |
-
max_tokens (int): Max tokens parameter.
|
94 |
-
temperature (float): Temperature parameter.
|
95 |
-
top_p (float): Top-p parameter.
|
96 |
-
|
97 |
-
Yields:
|
98 |
-
tuple: (updated chat history, cleared user input)
|
99 |
-
"""
|
100 |
for updated_history in respond(message, history, model_name, max_tokens, temperature, top_p):
|
101 |
yield updated_history, ""
|
102 |
|
103 |
-
def
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
|
114 |
-
return model_name
|
115 |
|
116 |
-
# Create the Gradio interface with Blocks
|
117 |
with gr.Blocks(css=".gr-button {margin: 5px; width: 100%;} .gr-column {padding: 10px;}") as demo:
|
118 |
-
# Title and description
|
119 |
gr.Markdown("# LeCarnet")
|
120 |
-
gr.Markdown("Select a model on the right and type a message to chat.")
|
121 |
-
|
122 |
-
# Two-column layout with specific widths
|
123 |
with gr.Row():
|
124 |
-
# Left column: Chat interface (80% width)
|
125 |
with gr.Column(scale=4):
|
|
|
126 |
chatbot = gr.Chatbot(
|
127 |
-
avatar_images=(None,
|
128 |
label="Chat",
|
129 |
-
height=600,
|
130 |
)
|
131 |
user_input = gr.Textbox(placeholder="Type your message here...", label="Message")
|
132 |
submit_btn = gr.Button("Send")
|
133 |
-
|
134 |
-
# Right column: Model selection and parameters (20% width)
|
135 |
with gr.Column(scale=1, min_width=200):
|
136 |
-
|
137 |
-
|
138 |
-
|
139 |
-
|
140 |
-
|
141 |
-
btn_3m = gr.Button("LeCarnet-3M")
|
142 |
-
btn_8m = gr.Button("LeCarnet-8M")
|
143 |
-
btn_21m = gr.Button("LeCarnet-21M")
|
144 |
-
|
145 |
-
# Sliders for parameters
|
146 |
max_tokens = gr.Slider(1, 512, value=512, step=1, label="Max New Tokens")
|
147 |
temperature = gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature")
|
148 |
top_p = gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p")
|
149 |
-
|
150 |
-
# Example prompts
|
151 |
-
examples = gr.Examples(
|
152 |
-
examples=[
|
153 |
-
["Il était une fois un petit garçon qui vivait dans un village paisible."],
|
154 |
-
["Il était une fois une grenouille qui rêvait de toucher les étoiles chaque nuit depuis son étang."],
|
155 |
-
["Il était une fois un petit lapin perdu"],
|
156 |
-
],
|
157 |
-
inputs=user_input,
|
158 |
-
)
|
159 |
-
|
160 |
-
# Event handling for submit button
|
161 |
submit_btn.click(
|
162 |
fn=submit,
|
163 |
-
inputs=[user_input, chatbot,
|
164 |
outputs=[chatbot, user_input],
|
165 |
)
|
166 |
-
|
167 |
-
|
168 |
-
|
169 |
-
|
170 |
-
inputs=[gr.State("LeCarnet-3M"), model_state],
|
171 |
-
outputs=model_state,
|
172 |
-
)
|
173 |
-
btn_8m.click(
|
174 |
-
fn=select_model,
|
175 |
-
inputs=[gr.State("LeCarnet-8M"), model_state],
|
176 |
-
outputs=model_state,
|
177 |
-
)
|
178 |
-
btn_21m.click(
|
179 |
-
fn=select_model,
|
180 |
-
inputs=[gr.State("LeCarnet-21M"), model_state],
|
181 |
-
outputs=model_state,
|
182 |
)
|
183 |
|
184 |
if __name__ == "__main__":
|
|
|
1 |
import os
|
2 |
import threading
|
3 |
from collections import defaultdict
|
4 |
+
from PIL import Image
|
5 |
+
import tempfile
|
6 |
import gradio as gr
|
7 |
from transformers import (
|
8 |
AutoModelForCausalLM,
|
|
|
10 |
TextIteratorStreamer,
|
11 |
)
|
12 |
|
|
|
13 |
model_name_to_path = {
|
14 |
"LeCarnet-3M": "MaxLSB/LeCarnet-3M",
|
15 |
"LeCarnet-8M": "MaxLSB/LeCarnet-8M",
|
16 |
"LeCarnet-21M": "MaxLSB/LeCarnet-21M",
|
17 |
}
|
18 |
|
|
|
19 |
hf_token = os.environ["HUGGINGFACEHUB_API_TOKEN"]
|
20 |
|
|
|
21 |
loaded_models = defaultdict(dict)
|
22 |
|
23 |
for name, path in model_name_to_path.items():
|
|
|
25 |
loaded_models[name]["model"] = AutoModelForCausalLM.from_pretrained(path, token=hf_token)
|
26 |
loaded_models[name]["model"].eval()
|
27 |
|
28 |
+
def resize_logo(input_path, size=(100, 100)):
|
29 |
+
with Image.open(input_path) as img:
|
30 |
+
img = img.resize(size, Image.LANCZOS)
|
31 |
+
temp_file = tempfile.NamedTemporaryFile(suffix=".png", delete=False)
|
32 |
+
img.save(temp_file.name, format="PNG")
|
33 |
+
return temp_file.name
|
34 |
+
|
35 |
def respond(message, history, model_name, max_tokens, temperature, top_p):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
history = history + [(message, "")]
|
37 |
+
yield history
|
|
|
|
|
38 |
tokenizer = loaded_models[model_name]["tokenizer"]
|
39 |
model = loaded_models[model_name]["model"]
|
|
|
|
|
40 |
inputs = tokenizer(message, return_tensors="pt")
|
|
|
|
|
41 |
streamer = TextIteratorStreamer(
|
42 |
tokenizer,
|
43 |
skip_prompt=False,
|
44 |
skip_special_tokens=True,
|
45 |
)
|
|
|
|
|
46 |
generate_kwargs = dict(
|
47 |
**inputs,
|
48 |
streamer=streamer,
|
|
|
52 |
top_p=top_p,
|
53 |
eos_token_id=tokenizer.eos_token_id,
|
54 |
)
|
|
|
|
|
55 |
thread = threading.Thread(target=model.generate, kwargs=generate_kwargs)
|
56 |
thread.start()
|
57 |
+
accumulated = f"**{model_name}**\n\n"
|
|
|
|
|
58 |
for new_text in streamer:
|
59 |
accumulated += new_text
|
60 |
history[-1] = (message, accumulated)
|
61 |
yield history
|
62 |
|
63 |
def submit(message, history, model_name, max_tokens, temperature, top_p):
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
64 |
for updated_history in respond(message, history, model_name, max_tokens, temperature, top_p):
|
65 |
yield updated_history, ""
|
66 |
|
67 |
+
def start_with_example(example, model_name, max_tokens, temperature, top_p):
|
68 |
+
for updated_history in respond(example, [], model_name, max_tokens, temperature, top_p):
|
69 |
+
yield updated_history, ""
|
70 |
+
|
71 |
+
resized_logo_path = resize_logo("media/le-carnet.png", size=(100, 100))
|
72 |
+
|
73 |
+
examples = [
|
74 |
+
"Il était une fois un petit garçon qui vivait dans un village paisible.",
|
75 |
+
"Il était une fois une grenouille qui rêvait de toucher les étoiles chaque nuit depuis son étang.",
|
76 |
+
"Il était une fois un petit lapin perdu",
|
77 |
+
]
|
|
|
78 |
|
|
|
79 |
with gr.Blocks(css=".gr-button {margin: 5px; width: 100%;} .gr-column {padding: 10px;}") as demo:
|
|
|
80 |
gr.Markdown("# LeCarnet")
|
81 |
+
gr.Markdown("Select a model on the right and type a message to chat, or choose an example below.")
|
|
|
|
|
82 |
with gr.Row():
|
|
|
83 |
with gr.Column(scale=4):
|
84 |
+
dataset = gr.Dataset(components=[gr.Textbox(visible=False)], samples=[[ex] for ex in examples], type="values")
|
85 |
chatbot = gr.Chatbot(
|
86 |
+
avatar_images=(None, resized_logo_path),
|
87 |
label="Chat",
|
88 |
+
height=600,
|
89 |
)
|
90 |
user_input = gr.Textbox(placeholder="Type your message here...", label="Message")
|
91 |
submit_btn = gr.Button("Send")
|
|
|
|
|
92 |
with gr.Column(scale=1, min_width=200):
|
93 |
+
model_dropdown = gr.Dropdown(
|
94 |
+
choices=list(model_name_to_path.keys()),
|
95 |
+
value="LeCarnet-8M",
|
96 |
+
label="Model"
|
97 |
+
)
|
|
|
|
|
|
|
|
|
|
|
98 |
max_tokens = gr.Slider(1, 512, value=512, step=1, label="Max New Tokens")
|
99 |
temperature = gr.Slider(0.1, 2.0, value=0.7, step=0.1, label="Temperature")
|
100 |
top_p = gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
submit_btn.click(
|
102 |
fn=submit,
|
103 |
+
inputs=[user_input, chatbot, model_dropdown, max_tokens, temperature, top_p],
|
104 |
outputs=[chatbot, user_input],
|
105 |
)
|
106 |
+
dataset.change(
|
107 |
+
fn=start_with_example,
|
108 |
+
inputs=[dataset, model_dropdown, max_tokens, temperature, top_p],
|
109 |
+
outputs=[chatbot, user_input],
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
110 |
)
|
111 |
|
112 |
if __name__ == "__main__":
|