Spaces:
Runtime error
Runtime error
Update app.py
Browse files
app.py
CHANGED
@@ -1,64 +1,65 @@
|
|
1 |
#!/usr/bin/env python
|
2 |
import gradio as gr
|
3 |
-
from huggingface_hub import InferenceClient
|
4 |
import spaces
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
|
|
|
|
|
|
9 |
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
|
18 |
-
):
|
19 |
-
messages = [{"role": "system", "content": system_message}]
|
20 |
|
21 |
-
for val in history:
|
22 |
-
if val[0]:
|
23 |
-
messages.append({"role": "user", "content": val[0]})
|
24 |
-
if val[1]:
|
25 |
-
messages.append({"role": "assistant", "content": val[1]})
|
26 |
|
27 |
-
|
28 |
|
29 |
-
|
|
|
|
|
30 |
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
|
40 |
-
|
41 |
-
|
|
|
|
|
|
|
|
|
|
|
42 |
|
43 |
-
"""
|
44 |
-
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
45 |
-
"""
|
46 |
-
demo = gr.ChatInterface(
|
47 |
-
respond,
|
48 |
-
additional_inputs=[
|
49 |
-
gr.Textbox(value="Your name is Aether, a 7 billion parameter large language model, you are a helpful and friendly chat assistant", label="System message"),
|
50 |
-
gr.Slider(minimum=1, maximum=4096, value=512, step=1, label="Max new tokens"),
|
51 |
-
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
52 |
-
gr.Slider(
|
53 |
-
minimum=0.1,
|
54 |
-
maximum=1.0,
|
55 |
-
value=0.95,
|
56 |
-
step=0.05,
|
57 |
-
label="Top-p (nucleus sampling)",
|
58 |
-
),
|
59 |
-
],
|
60 |
-
)
|
61 |
|
62 |
|
63 |
-
|
64 |
-
|
|
|
1 |
#!/usr/bin/env python
|
2 |
import gradio as gr
|
|
|
3 |
import spaces
|
4 |
+
import torch
|
5 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer
|
6 |
+
import time
|
7 |
+
import numpy as np
|
8 |
+
from torch.nn import functional as F
|
9 |
+
import os
|
10 |
+
from threading import Thread
|
11 |
|
12 |
+
print(f"Starting to load the model to memory")
|
13 |
+
m = AutoModelForCausalLM.from_pretrained(
|
14 |
+
"xi0v/aether-7b-chat-v1.0", torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, trust_remote_code=False)
|
15 |
+
tok = AutoTokenizer.from_pretrained("xi0v/aether-7b-chat-v1.0", trust_remote_code=False)
|
16 |
+
# using CUDA for an optimal experience
|
17 |
+
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
|
18 |
+
m = m.to(device)
|
19 |
+
print(f"Sucessfully loaded the model to the memory")
|
|
|
|
|
20 |
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
+
start_message = ""
|
23 |
|
24 |
+
def user(message, history):
|
25 |
+
# Append the user's message to the conversation history
|
26 |
+
return "", history + [[message, ""]]
|
27 |
|
28 |
+
@spaces.GPU
|
29 |
+
def chat(message, history):
|
30 |
+
chat = []
|
31 |
+
for item in history:
|
32 |
+
chat.append({"role": "user", "content": item[0]})
|
33 |
+
if item[1] is not None:
|
34 |
+
chat.append({"role": "assistant", "content": item[1]})
|
35 |
+
chat.append({"role": "user", "content": message})
|
36 |
+
messages = tok.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
|
37 |
+
# Tokenize the messages string
|
38 |
+
model_inputs = tok([messages], return_tensors="pt").to(device)
|
39 |
+
streamer = TextIteratorStreamer(
|
40 |
+
tok, timeout=10., skip_prompt=True, skip_special_tokens=True)
|
41 |
+
generate_kwargs = dict(
|
42 |
+
model_inputs,
|
43 |
+
streamer=streamer,
|
44 |
+
max_new_tokens=1024,
|
45 |
+
do_sample=True,
|
46 |
+
top_p=0.95,
|
47 |
+
top_k=1000,
|
48 |
+
temperature=0.75,
|
49 |
+
num_beams=1,
|
50 |
+
)
|
51 |
+
t = Thread(target=m.generate, kwargs=generate_kwargs)
|
52 |
+
t.start()
|
53 |
|
54 |
+
# Initialize an empty string to store the generated text
|
55 |
+
partial_text = ""
|
56 |
+
for new_text in streamer:
|
57 |
+
# print(new_text)
|
58 |
+
partial_text += new_text
|
59 |
+
# Yield an empty string to cleanup the message textbox and the updated conversation history
|
60 |
+
yield partial_text
|
61 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
62 |
|
63 |
|
64 |
+
demo = gr.ChatInterface(fn=chat, examples=["hello", "hola", "merhaba"], title="Stable LM 2 Zephyr 1.6b")
|
65 |
+
demo.launch()
|