xueshugpt / request_llm /bridge_all.py
qingxu99's picture
Rebase v3.0
91609d6
raw
history blame
5.47 kB
"""
该文件中主要包含2个函数
不具备多线程能力的函数:
1. predict: 正常对话时使用,具备完备的交互功能,不可多线程
具备多线程调用能力的函数
2. predict_no_ui_long_connection:在实验过程中发现调用predict_no_ui处理长文档时,和openai的连接容易断掉,这个函数用stream的方式解决这个问题,同样支持多线程
"""
from concurrent.futures import ThreadPoolExecutor
from .bridge_chatgpt import predict_no_ui_long_connection as chatgpt_noui
from .bridge_chatgpt import predict as chatgpt_ui
from .bridge_chatglm import predict_no_ui_long_connection as chatglm_noui
from .bridge_chatglm import predict as chatglm_ui
from .bridge_tgui import predict_no_ui_long_connection as tgui_noui
from .bridge_tgui import predict as tgui_ui
methods = {
"openai-no-ui": chatgpt_noui,
"openai-ui": chatgpt_ui,
"chatglm-no-ui": chatglm_noui,
"chatglm-ui": chatglm_ui,
"tgui-no-ui": tgui_noui,
"tgui-ui": tgui_ui,
}
def predict_no_ui_long_connection(inputs, llm_kwargs, history, sys_prompt, observe_window, console_slience=False):
"""
发送至LLM,等待回复,一次性完成,不显示中间过程。但内部用stream的方法避免中途网线被掐。
inputs:
是本次问询的输入
sys_prompt:
系统静默prompt
llm_kwargs:
LLM的内部调优参数
history:
是之前的对话列表
observe_window = None:
用于负责跨越线程传递已经输出的部分,大部分时候仅仅为了fancy的视觉效果,留空即可。observe_window[0]:观测窗。observe_window[1]:看门狗
"""
import threading, time, copy
model = llm_kwargs['llm_model']
n_model = 1
if '&' not in model:
assert not model.startswith("tgui"), "TGUI不支持函数插件的实现"
# 如果只询问1个大语言模型:
if model.startswith('gpt'):
method = methods['openai-no-ui']
elif model == 'chatglm':
method = methods['chatglm-no-ui']
elif model.startswith('tgui'):
method = methods['tgui-no-ui']
return method(inputs, llm_kwargs, history, sys_prompt, observe_window, console_slience)
else:
# 如果同时询问多个大语言模型:
executor = ThreadPoolExecutor(max_workers=16)
models = model.split('&')
n_model = len(models)
window_len = len(observe_window)
if window_len==0:
window_mutex = [[] for _ in range(n_model)] + [True]
elif window_len==1:
window_mutex = [[""] for _ in range(n_model)] + [True]
elif window_len==2:
window_mutex = [["", time.time()] for _ in range(n_model)] + [True]
futures = []
for i in range(n_model):
model = models[i]
if model.startswith('gpt'):
method = methods['openai-no-ui']
elif model == 'chatglm':
method = methods['chatglm-no-ui']
elif model.startswith('tgui'):
method = methods['tgui-no-ui']
llm_kwargs_feedin = copy.deepcopy(llm_kwargs)
llm_kwargs_feedin['llm_model'] = model
future = executor.submit(method, inputs, llm_kwargs_feedin, history, sys_prompt, window_mutex[i], console_slience)
futures.append(future)
def mutex_manager(window_mutex, observe_window):
while True:
time.sleep(0.2)
if not window_mutex[-1]: break
# 看门狗(watchdog)
for i in range(n_model):
window_mutex[i][1] = observe_window[1]
# 观察窗(window)
chat_string = []
for i in range(n_model):
chat_string.append( f"[{str(models[i])} 说]: {window_mutex[i][0]}" )
res = '\n\n---\n\n'.join(chat_string)
# # # # # # # # # # #
observe_window[0] = res
t_model = threading.Thread(target=mutex_manager, args=(window_mutex, observe_window), daemon=True)
t_model.start()
return_string_collect = []
for i, future in enumerate(futures): # wait and get
return_string_collect.append( f"[{str(models[i])} 说]: {future.result()}" )
window_mutex[-1] = False # stop mutex thread
res = '\n\n---\n\n'.join(return_string_collect)
return res
def predict(inputs, llm_kwargs, *args, **kwargs):
"""
发送至LLM,流式获取输出。
用于基础的对话功能。
inputs 是本次问询的输入
top_p, temperature是LLM的内部调优参数
history 是之前的对话列表(注意无论是inputs还是history,内容太长了都会触发token数量溢出的错误)
chatbot 为WebUI中显示的对话列表,修改它,然后yeild出去,可以直接修改对话界面内容
additional_fn代表点击的哪个按钮,按钮见functional.py
"""
if llm_kwargs['llm_model'].startswith('gpt'):
method = methods['openai-ui']
elif llm_kwargs['llm_model'] == 'chatglm':
method = methods['chatglm-ui']
elif llm_kwargs['llm_model'].startswith('tgui'):
method = methods['tgui-ui']
yield from method(inputs, llm_kwargs, *args, **kwargs)