File size: 5,332 Bytes
7cfa5bf 8072750 dbdc900 8072750 dbdc900 7cfa5bf 70d0b73 7cfa5bf 8072750 7cfa5bf 8072750 70d0b73 8072750 7cfa5bf 8072750 70d0b73 8072750 a1a8972 aec6f8a 70d0b73 aec6f8a 7fb0028 8d28fe5 7fb0028 e337119 70d0b73 0c41d6c 70d0b73 8072750 70d0b73 7cfa5bf 8072750 7cfa5bf dbdc900 7cfa5bf |
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 |
import gradio as gr
import requests
import json
import os
from dotenv import load_dotenv
load_dotenv()
API_URL = os.getenv("API_URL")
API_TOKEN = os.getenv("API_TOKEN")
if not API_URL or not API_TOKEN:
raise ValueError("invalid API_URL || API_TOKEN")
print(f"[INFO] starting:")
print(f"[INFO] API_URL: {API_URL[:6]}...{API_URL[-12:]}")
print(f"[INFO] API_TOKEN: {API_TOKEN[:10]}...{API_TOKEN[-10:]}")
"""
For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
"""
def respond(
message,
history: list[dict],
system_message,
max_tokens,
temperature,
top_p,
):
messages = [{"role": "system", "content": system_message}]
messages.extend(history)
messages.append({"role": "user", "content": message})
headers = {
"Content-Type": "application/json",
"Authorization": f"Bearer {API_TOKEN}"
}
data = {
"model": "/data/DMind-1",
"stream": True,
"messages": messages,
"temperature": temperature,
"top_p": top_p,
"top_k": 20,
"min_p": 0.1,
"max_tokens": 32768
}
try:
with requests.post(API_URL, headers=headers, json=data, stream=True) as r:
if r.status_code == 200:
current_response = ""
buffer = "" # 用于累积可能被分割的内容
in_think_block = False
for line in r.iter_lines():
if line:
line = line.decode('utf-8')
if line.startswith('data: '):
try:
json_response = json.loads(line[6:])
if 'choices' in json_response and len(json_response['choices']) > 0:
delta = json_response['choices'][0].get('delta', {})
if 'content' in delta:
content = delta['content']
if content:
buffer += content
if not in_think_block and '<think>' in buffer:
in_think_block = True
parts = buffer.split('<think>')
if len(parts) > 1:
if parts[0]:
current_response += parts[0]
current_response += '```\n'
buffer = parts[1]
if in_think_block and ('</think>' in buffer or '**Final Answer**' in buffer):
in_think_block = False
current_response += buffer.split('</think>')[0] + '\n```'
if '</think>' in buffer:
buffer = buffer.split('</think>')[1]
else:
buffer = buffer.split('**Final Answer**')[1]
if not in_think_block and buffer:
current_response += buffer
buffer = ""
formatted_response = current_response
formatted_response = formatted_response.replace('<', '<').replace('>', '>')
formatted_response = formatted_response.replace('*', '\\*')
yield formatted_response
except json.JSONDecodeError:
continue
else:
print(f"[ERROR] Bad status code: {r.status_code}, response: {r.text}")
yield "Service temporarily unavailable"
except Exception as e:
print(f"[ERROR] Request error: {e}")
yield "Service error occurred"
"""
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
"""
demo = gr.ChatInterface(
respond,
additional_inputs=[
gr.Textbox(value="You are DMind AI Assistant, built by DMind.AI. Never present as Qwen, ChatGPT, Tongyi, OpenAI, etc.", label="System message"),
gr.Slider(minimum=1, maximum=32768, value=16384, step=1, label="Max new tokens"),
gr.Slider(minimum=0.1, maximum=4.0, value=0.6, step=0.1, label="Temperature"),
gr.Slider(
minimum=0.1,
maximum=1.0,
value=0.95,
step=0.05,
label="Top-p (nucleus sampling)",
),
],
type="messages"
)
if __name__ == "__main__":
demo.launch()
|