CC/gpt_academic
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

update gr

Browse Source
This commit is contained in:
binary-husky
2024-01-23 15:54:09 +08:00
parent eaf27df32a
commit 47289f863d
85 changed files with 5302 additions and 1788 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
request_llms/README.md
View File

@@ -32,4 +32,4 @@ P.S. 如果您按照以下步骤成功接入了新的大模型，欢迎发Pull R
5. 测试通过后，在`request_llms/bridge_all.py`中做最后的修改，把你的模型完全接入到框架中（聪慧如您，只需要看一眼该文件就明白怎么修改了）
6. 修改`LLM_MODEL`配置，然后运行`python main.py`，测试最后的效果
6. 修改`LLM_MODEL`配置，然后运行`python main.py`，测试最后的效果

52
request_llms/bridge_all.py
View File

@@ -28,6 +28,9 @@ from .bridge_chatglm3 import predict as chatglm3_ui
from .bridge_qianfan import predict_no_ui_long_connection as qianfan_noui
from .bridge_qianfan import predict as qianfan_ui
from .bridge_google_gemini import predict as genai_ui
from .bridge_google_gemini import predict_no_ui_long_connection as genai_noui
colors = ['#FF00FF', '#00FFFF', '#FF0000', '#990099', '#009999', '#990044']
class LazyloadTiktoken(object):
@@ -246,6 +249,22 @@ model_info = {
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
"gemini-pro": {
"fn_with_ui": genai_ui,
"fn_without_ui": genai_noui,
"endpoint": None,
"max_token": 1024 * 32,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
"gemini-pro-vision": {
"fn_with_ui": genai_ui,
"fn_without_ui": genai_noui,
"endpoint": None,
"max_token": 1024 * 32,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
},
}
# -=-=-=-=-=-=- api2d 对齐支持 -=-=-=-=-=-=-
@@ -479,22 +498,6 @@ if "qwen-turbo" in AVAIL_LLM_MODELS or "qwen-plus" in AVAIL_LLM_MODELS or "qwen-
})
except:
print(trimmed_format_exc())
if "chatgpt_website" in AVAIL_LLM_MODELS: # 接入一些逆向工程https://github.com/acheong08/ChatGPT-to-API/
try:
from .bridge_chatgpt_website import predict_no_ui_long_connection as chatgpt_website_noui
from .bridge_chatgpt_website import predict as chatgpt_website_ui
model_info.update({
"chatgpt_website": {
"fn_with_ui": chatgpt_website_ui,
"fn_without_ui": chatgpt_website_noui,
"endpoint": openai_endpoint,
"max_token": 4096,
"tokenizer": tokenizer_gpt35,
"token_cnt": get_token_num_gpt35,
}
})
except:
print(trimmed_format_exc())
if "spark" in AVAIL_LLM_MODELS: # 讯飞星火认知大模型
try:
from .bridge_spark import predict_no_ui_long_connection as spark_noui
@@ -591,6 +594,23 @@ if "deepseekcoder" in AVAIL_LLM_MODELS: # deepseekcoder
})
except:
print(trimmed_format_exc())
# if "skylark" in AVAIL_LLM_MODELS:
# try:
# from .bridge_skylark2 import predict_no_ui_long_connection as skylark_noui
# from .bridge_skylark2 import predict as skylark_ui
# model_info.update({
# "skylark": {
# "fn_with_ui": skylark_ui,
# "fn_without_ui": skylark_noui,
# "endpoint": None,
# "max_token": 4096,
# "tokenizer": tokenizer_gpt35,
# "token_cnt": get_token_num_gpt35,
# }
# })
# except:
# print(trimmed_format_exc())
# <-- 用于定义和切换多个azure模型 -->
AZURE_CFG_ARRAY = get_conf("AZURE_CFG_ARRAY")

3
request_llms/bridge_chatgpt.py
View File

@@ -244,6 +244,9 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
if has_choices and not choice_valid:
# 一些垃圾第三方接口的出现这样的错误
continue
if ('data: [DONE]' not in chunk_decoded) and len(chunk_decoded) > 0 and (chunkjson is None):
# 传递进来一些奇怪的东西
raise ValueError(f'无法读取以下数据，请检查配置。\n\n{chunk_decoded}')
# 前者是API2D的结束条件，后者是OPENAI的结束条件
if ('data: [DONE]' in chunk_decoded) or (len(chunkjson['choices'][0]["delta"]) == 0):
# 判定为数据流的结束，gpt_replying_buffer也写完了

114
request_llms/bridge_google_gemini.py Normal file
View File

@@ -0,0 +1,114 @@
# encoding: utf-8
# @Time : 2023/12/21
# @Author : Spike
# @Descr :
import json
import re
import os
import time
from request_llms.com_google import GoogleChatInit
from toolbox import get_conf, update_ui, update_ui_lastest_msg, have_any_recent_upload_image_files, trimmed_format_exc
proxies, TIMEOUT_SECONDS, MAX_RETRY = get_conf('proxies', 'TIMEOUT_SECONDS', 'MAX_RETRY')
timeout_bot_msg = '[Local Message] Request timeout. Network error. Please check proxy settings in config.py.' + \
'网络错误，检查代理服务器是否可用，以及代理设置的格式是否正确，格式须是[协议]://[地址]:[端口]，缺一不可。'
def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="", observe_window=None,
console_slience=False):
# 检查API_KEY
if get_conf("GEMINI_API_KEY") == "":
raise ValueError(f"请配置 GEMINI_API_KEY。")
genai = GoogleChatInit()
watch_dog_patience = 5 # 看门狗的耐心, 设置5秒即可
gpt_replying_buffer = ''
stream_response = genai.generate_chat(inputs, llm_kwargs, history, sys_prompt)
for response in stream_response:
results = response.decode()
match = re.search(r'"text":\s*"((?:[^"\\]|\\.)*)"', results, flags=re.DOTALL)
error_match = re.search(r'\"message\":\s*\"(.*?)\"', results, flags=re.DOTALL)
if match:
try:
paraphrase = json.loads('{"text": "%s"}' % match.group(1))
except:
raise ValueError(f"解析GEMINI消息出错。")
buffer = paraphrase['text']
gpt_replying_buffer += buffer
if len(observe_window) >= 1:
observe_window[0] = gpt_replying_buffer
if len(observe_window) >= 2:
if (time.time() - observe_window[1]) > watch_dog_patience: raise RuntimeError("程序终止。")
if error_match:
raise RuntimeError(f'{gpt_replying_buffer} 对话错误')
return gpt_replying_buffer
def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream=True, additional_fn=None):
# 检查API_KEY
if get_conf("GEMINI_API_KEY") == "":
yield from update_ui_lastest_msg(f"请配置 GEMINI_API_KEY。", chatbot=chatbot, history=history, delay=0)
return
# 适配润色区域
if additional_fn is not None:
from core_functional import handle_core_functionality
inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
if "vision" in llm_kwargs["llm_model"]:
have_recent_file, image_paths = have_any_recent_upload_image_files(chatbot)
def make_media_input(inputs, image_paths):
for image_path in image_paths:
inputs = inputs + f'<br/><br/><div align="center"><img src="file={os.path.abspath(image_path)}"></div>'
return inputs
if have_recent_file:
inputs = make_media_input(inputs, image_paths)
chatbot.append((inputs, ""))
yield from update_ui(chatbot=chatbot, history=history)
genai = GoogleChatInit()
retry = 0
while True:
try:
stream_response = genai.generate_chat(inputs, llm_kwargs, history, system_prompt)
break
except Exception as e:
retry += 1
chatbot[-1] = ((chatbot[-1][0], trimmed_format_exc()))
yield from update_ui(chatbot=chatbot, history=history, msg="请求失败") # 刷新界面
return
gpt_replying_buffer = ""
gpt_security_policy = ""
history.extend([inputs, ''])
for response in stream_response:
results = response.decode("utf-8") # 被这个解码给耍了。。
gpt_security_policy += results
match = re.search(r'"text":\s*"((?:[^"\\]|\\.)*)"', results, flags=re.DOTALL)
error_match = re.search(r'\"message\":\s*\"(.*)\"', results, flags=re.DOTALL)
if match:
try:
paraphrase = json.loads('{"text": "%s"}' % match.group(1))
except:
raise ValueError(f"解析GEMINI消息出错。")
gpt_replying_buffer += paraphrase['text'] # 使用 json 解析库进行处理
chatbot[-1] = (inputs, gpt_replying_buffer)
history[-1] = gpt_replying_buffer
yield from update_ui(chatbot=chatbot, history=history)
if error_match:
history = history[-2] # 错误的不纳入对话
chatbot[-1] = (inputs, gpt_replying_buffer + f"对话错误，请查看message\n\n```\n{error_match.group(1)}\n```")
yield from update_ui(chatbot=chatbot, history=history)
raise RuntimeError('对话错误')
if not gpt_replying_buffer:
history = history[-2] # 错误的不纳入对话
chatbot[-1] = (inputs, gpt_replying_buffer + f"触发了Google的安全访问策略，没有回答\n\n```\n{gpt_security_policy}\n```")
yield from update_ui(chatbot=chatbot, history=history)
if __name__ == '__main__':
import sys
llm_kwargs = {'llm_model': 'gemini-pro'}
result = predict('Write long a story about a magic backpack.', llm_kwargs, llm_kwargs, [])
for i in result:
print(i)

224
request_llms/bridge_newbingfree.py
View File

@@ -1,16 +1,17 @@
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第一部分：来自EdgeGPT.py
https://github.com/acheong08/EdgeGPT
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
from .edge_gpt_free import Chatbot as NewbingChatbot
load_message = "等待NewBing响应。"
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第二部分：子进程Worker（调用主体）
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
import time
import json
@@ -22,19 +23,30 @@ import threading
from toolbox import update_ui, get_conf, trimmed_format_exc
from multiprocessing import Process, Pipe
def preprocess_newbing_out(s):
pattern = r'\^(\d+)\^' # 匹配^数字^
sub = lambda m: '('+m.group(1)+')' # 将匹配到的数字作为替换值
result = re.sub(pattern, sub, s) # 替换操作
if '[1]' in result:
result += '\n\n```reference\n' + "\n".join([r for r in result.split('\n') if r.startswith('[')]) + '\n```\n'
pattern = r"\^(\d+)\^" # 匹配^数字^
sub = lambda m: "(" + m.group(1) + ")" # 将匹配到的数字作为替换值
result = re.sub(pattern, sub, s) # 替换操作
if "[1]" in result:
result += (
"\n\n```reference\n"
+ "\n".join([r for r in result.split("\n") if r.startswith("[")])
+ "\n```\n"
)
return result
def preprocess_newbing_out_simple(result):
if '[1]' in result:
result += '\n\n```reference\n' + "\n".join([r for r in result.split('\n') if r.startswith('[')]) + '\n```\n'
if "[1]" in result:
result += (
"\n\n```reference\n"
+ "\n".join([r for r in result.split("\n") if r.startswith("[")])
+ "\n```\n"
)
return result
class NewBingHandle(Process):
def __init__(self):
super().__init__(daemon=True)
@@ -46,11 +58,12 @@ class NewBingHandle(Process):
self.check_dependency()
self.start()
self.threadLock = threading.Lock()
def check_dependency(self):
try:
self.success = False
import certifi, httpx, rich
self.info = "依赖检测通过，等待NewBing响应。注意目前不能多人同时调用NewBing接口（有线程锁），否则将导致每个人的NewBing问询历史互相渗透。调用NewBing时，会自动使用已配置的代理。"
self.success = True
except:
@@ -62,18 +75,19 @@ class NewBingHandle(Process):
async def async_run(self):
# 读取配置
NEWBING_STYLE = get_conf('NEWBING_STYLE')
NEWBING_STYLE = get_conf("NEWBING_STYLE")
from request_llms.bridge_all import model_info
endpoint = model_info['newbing']['endpoint']
endpoint = model_info["newbing"]["endpoint"]
while True:
# 等待
kwargs = self.child.recv()
question=kwargs['query']
history=kwargs['history']
system_prompt=kwargs['system_prompt']
question = kwargs["query"]
history = kwargs["history"]
system_prompt = kwargs["system_prompt"]
# 是否重置
if len(self.local_history) > 0 and len(history)==0:
if len(self.local_history) > 0 and len(history) == 0:
await self.newbing_model.reset()
self.local_history = []
@@ -81,34 +95,33 @@ class NewBingHandle(Process):
prompt = ""
if system_prompt not in self.local_history:
self.local_history.append(system_prompt)
prompt += system_prompt + '\n'
prompt += system_prompt + "\n"
# 追加历史
for ab in history:
a, b = ab
if a not in self.local_history:
self.local_history.append(a)
prompt += a + '\n'
prompt += a + "\n"
# 问题
prompt += question
self.local_history.append(question)
print('question:', prompt)
print("question:", prompt)
# 提交
async for final, response in self.newbing_model.ask_stream(
prompt=question,
conversation_style=NEWBING_STYLE, # ["creative", "balanced", "precise"]
wss_link=endpoint, # "wss://sydney.bing.com/sydney/ChatHub"
conversation_style=NEWBING_STYLE, # ["creative", "balanced", "precise"]
wss_link=endpoint, # "wss://sydney.bing.com/sydney/ChatHub"
):
if not final:
print(response)
self.child.send(str(response))
else:
print('-------- receive final ---------')
self.child.send('[Finish]')
print("-------- receive final ---------")
self.child.send("[Finish]")
# self.local_history.append(response)
def run(self):
"""
这个函数运行在子进程
@@ -118,32 +131,37 @@ class NewBingHandle(Process):
self.local_history = []
if (self.newbing_model is None) or (not self.success):
# 代理设置
proxies, NEWBING_COOKIES = get_conf('proxies', 'NEWBING_COOKIES')
if proxies is None:
proxies, NEWBING_COOKIES = get_conf("proxies", "NEWBING_COOKIES")
if proxies is None:
self.proxies_https = None
else:
self.proxies_https = proxies['https']
else:
self.proxies_https = proxies["https"]
if (NEWBING_COOKIES is not None) and len(NEWBING_COOKIES) > 100:
try:
cookies = json.loads(NEWBING_COOKIES)
except:
self.success = False
tb_str = '\n```\n' + trimmed_format_exc() + '\n```\n'
self.child.send(f'[Local Message] NEWBING_COOKIES未填写或有格式错误。')
self.child.send('[Fail]'); self.child.send('[Finish]')
tb_str = "\n```\n" + trimmed_format_exc() + "\n```\n"
self.child.send(f"[Local Message] NEWBING_COOKIES未填写或有格式错误。")
self.child.send("[Fail]")
self.child.send("[Finish]")
raise RuntimeError(f"NEWBING_COOKIES未填写或有格式错误。")
else:
cookies = None
try:
self.newbing_model = NewbingChatbot(proxy=self.proxies_https, cookies=cookies)
self.newbing_model = NewbingChatbot(
proxy=self.proxies_https, cookies=cookies
)
except:
self.success = False
tb_str = '\n```\n' + trimmed_format_exc() + '\n```\n'
self.child.send(f'[Local Message] 不能加载Newbing组件，请注意Newbing组件已不再维护。{tb_str}')
self.child.send('[Fail]')
self.child.send('[Finish]')
tb_str = "\n```\n" + trimmed_format_exc() + "\n```\n"
self.child.send(
f"[Local Message] 不能加载Newbing组件，请注意Newbing组件已不再维护。{tb_str}"
)
self.child.send("[Fail]")
self.child.send("[Finish]")
raise RuntimeError(f"不能加载Newbing组件，请注意Newbing组件已不再维护。")
self.success = True
@@ -151,66 +169,100 @@ class NewBingHandle(Process):
# 进入任务等待状态
asyncio.run(self.async_run())
except Exception:
tb_str = '\n```\n' + trimmed_format_exc() + '\n```\n'
self.child.send(f'[Local Message] Newbing 请求失败，报错信息如下. 如果是与网络相关的问题，建议更换代理协议（推荐http）或代理节点 {tb_str}.')
self.child.send('[Fail]')
self.child.send('[Finish]')
tb_str = "\n```\n" + trimmed_format_exc() + "\n```\n"
self.child.send(
f"[Local Message] Newbing 请求失败，报错信息如下. 如果是与网络相关的问题，建议更换代理协议（推荐http）或代理节点 {tb_str}."
)
self.child.send("[Fail]")
self.child.send("[Finish]")
def stream_chat(self, **kwargs):
"""
这个函数运行在主进程
"""
self.threadLock.acquire() # 获取线程锁
self.parent.send(kwargs) # 请求子进程
self.threadLock.acquire() # 获取线程锁
self.parent.send(kwargs) # 请求子进程
while True:
res = self.parent.recv() # 等待newbing回复的片段
if res == '[Finish]': break # 结束
elif res == '[Fail]': self.success = False; break # 失败
else: yield res # newbing回复的片段
self.threadLock.release() # 释放线程锁
res = self.parent.recv() # 等待newbing回复的片段
if res == "[Finish]":
break # 结束
elif res == "[Fail]":
self.success = False
break # 失败
else:
yield res # newbing回复的片段
self.threadLock.release() # 释放线程锁
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第三部分：主进程统一调用函数接口
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
global newbingfree_handle
newbingfree_handle = None
def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="", observe_window=[], console_slience=False):
def predict_no_ui_long_connection(
inputs,
llm_kwargs,
history=[],
sys_prompt="",
observe_window=[],
console_slience=False,
):
"""
多线程方法
函数的说明请见 request_llms/bridge_all.py
多线程方法
函数的说明请见 request_llms/bridge_all.py
"""
global newbingfree_handle
if (newbingfree_handle is None) or (not newbingfree_handle.success):
newbingfree_handle = NewBingHandle()
if len(observe_window) >= 1: observe_window[0] = load_message + "\n\n" + newbingfree_handle.info
if not newbingfree_handle.success:
if len(observe_window) >= 1:
observe_window[0] = load_message + "\n\n" + newbingfree_handle.info
if not newbingfree_handle.success:
error = newbingfree_handle.info
newbingfree_handle = None
raise RuntimeError(error)
# 没有 sys_prompt 接口，因此把prompt加入 history
history_feedin = []
for i in range(len(history)//2):
history_feedin.append([history[2*i], history[2*i+1]] )
for i in range(len(history) // 2):
history_feedin.append([history[2 * i], history[2 * i + 1]])
watch_dog_patience = 5 # 看门狗 (watchdog) 的耐心, 设置5秒即可
watch_dog_patience = 5 # 看门狗 (watchdog) 的耐心, 设置5秒即可
response = ""
if len(observe_window) >= 1: observe_window[0] = "[Local Message] 等待NewBing响应中 ..."
for response in newbingfree_handle.stream_chat(query=inputs, history=history_feedin, system_prompt=sys_prompt, max_length=llm_kwargs['max_length'], top_p=llm_kwargs['top_p'], temperature=llm_kwargs['temperature']):
if len(observe_window) >= 1: observe_window[0] = preprocess_newbing_out_simple(response)
if len(observe_window) >= 2:
if (time.time()-observe_window[1]) > watch_dog_patience:
if len(observe_window) >= 1:
observe_window[0] = "[Local Message] 等待NewBing响应中 ..."
for response in newbingfree_handle.stream_chat(
query=inputs,
history=history_feedin,
system_prompt=sys_prompt,
max_length=llm_kwargs["max_length"],
top_p=llm_kwargs["top_p"],
temperature=llm_kwargs["temperature"],
):
if len(observe_window) >= 1:
observe_window[0] = preprocess_newbing_out_simple(response)
if len(observe_window) >= 2:
if (time.time() - observe_window[1]) > watch_dog_patience:
raise RuntimeError("程序终止。")
return preprocess_newbing_out_simple(response)
def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream = True, additional_fn=None):
def predict(
inputs,
llm_kwargs,
plugin_kwargs,
chatbot,
history=[],
system_prompt="",
stream=True,
additional_fn=None,
):
"""
单线程方法
函数的说明请见 request_llms/bridge_all.py
单线程方法
函数的说明请见 request_llms/bridge_all.py
"""
chatbot.append((inputs, "[Local Message] 等待NewBing响应中 ..."))
@@ -219,27 +271,41 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
newbingfree_handle = NewBingHandle()
chatbot[-1] = (inputs, load_message + "\n\n" + newbingfree_handle.info)
yield from update_ui(chatbot=chatbot, history=[])
if not newbingfree_handle.success:
if not newbingfree_handle.success:
newbingfree_handle = None
return
if additional_fn is not None:
from core_functional import handle_core_functionality
inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
inputs, history = handle_core_functionality(
additional_fn, inputs, history, chatbot
)
history_feedin = []
for i in range(len(history)//2):
history_feedin.append([history[2*i], history[2*i+1]] )
for i in range(len(history) // 2):
history_feedin.append([history[2 * i], history[2 * i + 1]])
chatbot[-1] = (inputs, "[Local Message] 等待NewBing响应中 ...")
response = "[Local Message] 等待NewBing响应中 ..."
yield from update_ui(chatbot=chatbot, history=history, msg="NewBing响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。")
for response in newbingfree_handle.stream_chat(query=inputs, history=history_feedin, system_prompt=system_prompt, max_length=llm_kwargs['max_length'], top_p=llm_kwargs['top_p'], temperature=llm_kwargs['temperature']):
yield from update_ui(
chatbot=chatbot, history=history, msg="NewBing响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。"
)
for response in newbingfree_handle.stream_chat(
query=inputs,
history=history_feedin,
system_prompt=system_prompt,
max_length=llm_kwargs["max_length"],
top_p=llm_kwargs["top_p"],
temperature=llm_kwargs["temperature"],
):
chatbot[-1] = (inputs, preprocess_newbing_out(response))
yield from update_ui(chatbot=chatbot, history=history, msg="NewBing响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。")
if response == "[Local Message] 等待NewBing响应中 ...": response = "[Local Message] NewBing响应异常，请刷新界面重试 ..."
yield from update_ui(
chatbot=chatbot, history=history, msg="NewBing响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。"
)
if response == "[Local Message] 等待NewBing响应中 ...":
response = "[Local Message] NewBing响应异常，请刷新界面重试 ..."
history.extend([inputs, response])
logging.info(f'[raw_input] {inputs}')
logging.info(f'[response] {response}')
logging.info(f"[raw_input] {inputs}")
logging.info(f"[response] {response}")
yield from update_ui(chatbot=chatbot, history=history, msg="完成全部响应，请提交新问题。")

67
request_llms/bridge_skylark2.py Normal file
View File

@@ -0,0 +1,67 @@
import time
from toolbox import update_ui, get_conf, update_ui_lastest_msg
from toolbox import check_packages, report_exception
model_name = '云雀大模型'
def validate_key():
YUNQUE_SECRET_KEY = get_conf("YUNQUE_SECRET_KEY")
if YUNQUE_SECRET_KEY == '': return False
return True
def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="", observe_window=[], console_slience=False):
"""
⭐ 多线程方法
函数的说明请见 request_llms/bridge_all.py
"""
watch_dog_patience = 5
response = ""
if validate_key() is False:
raise RuntimeError('请配置YUNQUE_SECRET_KEY')
from .com_skylark2api import YUNQUERequestInstance
sri = YUNQUERequestInstance()
for response in sri.generate(inputs, llm_kwargs, history, sys_prompt):
if len(observe_window) >= 1:
observe_window[0] = response
if len(observe_window) >= 2:
if (time.time()-observe_window[1]) > watch_dog_patience: raise RuntimeError("程序终止。")
return response
def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream = True, additional_fn=None):
"""
⭐ 单线程方法
函数的说明请见 request_llms/bridge_all.py
"""
chatbot.append((inputs, ""))
yield from update_ui(chatbot=chatbot, history=history)
# 尝试导入依赖，如果缺少依赖，则给出安装建议
try:
check_packages(["zhipuai"])
except:
yield from update_ui_lastest_msg(f"导入软件依赖失败。使用该模型需要额外依赖，安装方法```pip install --upgrade zhipuai```。",
chatbot=chatbot, history=history, delay=0)
return
if validate_key() is False:
yield from update_ui_lastest_msg(lastmsg="[Local Message] 请配置HUOSHAN_API_KEY", chatbot=chatbot, history=history, delay=0)
return
if additional_fn is not None:
from core_functional import handle_core_functionality
inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
# 开始接收回复
from .com_skylark2api import YUNQUERequestInstance
sri = YUNQUERequestInstance()
for response in sri.generate(inputs, llm_kwargs, history, system_prompt):
chatbot[-1] = (inputs, response)
yield from update_ui(chatbot=chatbot, history=history)
# 总结输出
if response == f"[Local Message] 等待{model_name}响应中 ...":
response = f"[Local Message] {model_name}响应异常 ..."
history.extend([inputs, response])
yield from update_ui(chatbot=chatbot, history=history)

185
request_llms/bridge_stackclaude.py
View File

@@ -7,14 +7,15 @@ import logging
import time
from toolbox import get_conf
import asyncio
load_message = "正在加载Claude组件，请稍候..."
try:
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第一部分：Slack API Client
https://github.com/yokonsan/claude-in-slack-api
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
from slack_sdk.errors import SlackApiError
@@ -23,20 +24,23 @@ try:
class SlackClient(AsyncWebClient):
"""SlackClient类用于与Slack API进行交互，实现消息发送、接收等功能。
属性：
- CHANNEL_ID：str类型，表示频道ID。
属性：
- CHANNEL_ID：str类型，表示频道ID。
方法：
- open_channel()：异步方法。通过调用conversations_open方法打开一个频道，并将返回的频道ID保存在属性CHANNEL_ID中。
- chat(text: str)：异步方法。向已打开的频道发送一条文本消息。
- get_slack_messages()：异步方法。获取已打开频道的最新消息并返回消息列表，目前不支持历史消息查询。
- get_reply()：异步方法。循环监听已打开频道的消息，如果收到"Typing…_"结尾的消息说明Claude还在继续输出，否则结束循环。
方法：
- open_channel()：异步方法。通过调用conversations_open方法打开一个频道，并将返回的频道ID保存在属性CHANNEL_ID中。
- chat(text: str)：异步方法。向已打开的频道发送一条文本消息。
- get_slack_messages()：异步方法。获取已打开频道的最新消息并返回消息列表，目前不支持历史消息查询。
- get_reply()：异步方法。循环监听已打开频道的消息，如果收到"Typing…_"结尾的消息说明Claude还在继续输出，否则结束循环。
"""
CHANNEL_ID = None
async def open_channel(self):
response = await self.conversations_open(users=get_conf('SLACK_CLAUDE_BOT_ID'))
response = await self.conversations_open(
users=get_conf("SLACK_CLAUDE_BOT_ID")
)
self.CHANNEL_ID = response["channel"]["id"]
async def chat(self, text):
@@ -49,33 +53,39 @@ try:
async def get_slack_messages(self):
try:
# TODO：暂时不支持历史消息，因为在同一个频道里存在多人使用时历史消息渗透问题
resp = await self.conversations_history(channel=self.CHANNEL_ID, oldest=self.LAST_TS, limit=1)
msg = [msg for msg in resp["messages"]
if msg.get("user") == get_conf('SLACK_CLAUDE_BOT_ID')]
resp = await self.conversations_history(
channel=self.CHANNEL_ID, oldest=self.LAST_TS, limit=1
)
msg = [
msg
for msg in resp["messages"]
if msg.get("user") == get_conf("SLACK_CLAUDE_BOT_ID")
]
return msg
except (SlackApiError, KeyError) as e:
raise RuntimeError(f"获取Slack消息失败。")
async def get_reply(self):
while True:
slack_msgs = await self.get_slack_messages()
if len(slack_msgs) == 0:
await asyncio.sleep(0.5)
continue
msg = slack_msgs[-1]
if msg["text"].endswith("Typing…_"):
yield False, msg["text"]
else:
yield True, msg["text"]
break
except:
pass
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第二部分：子进程Worker（调用主体）
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
@@ -88,7 +98,7 @@ class ClaudeHandle(Process):
self.success = True
self.local_history = []
self.check_dependency()
if self.success:
if self.success:
self.start()
self.threadLock = threading.Lock()
@@ -96,6 +106,7 @@ class ClaudeHandle(Process):
try:
self.success = False
import slack_sdk
self.info = "依赖检测通过，等待Claude响应。注意目前不能多人同时调用Claude接口（有线程锁），否则将导致每个人的Claude问询历史互相渗透。调用Claude时，会自动使用已配置的代理。"
self.success = True
except:
@@ -103,40 +114,44 @@ class ClaudeHandle(Process):
self.success = False
def ready(self):
return self.claude_model is not None
return self.claude_model is not None
async def async_run(self):
await self.claude_model.open_channel()
while True:
# 等待
kwargs = self.child.recv()
question = kwargs['query']
history = kwargs['history']
question = kwargs["query"]
history = kwargs["history"]
# 开始问问题
prompt = ""
# 问题
prompt += question
print('question:', prompt)
print("question:", prompt)
# 提交
await self.claude_model.chat(prompt)
# 获取回复
async for final, response in self.claude_model.get_reply():
async for final, response in self.claude_model.get_reply():
if not final:
print(response)
self.child.send(str(response))
else:
# 防止丢失最后一条消息
slack_msgs = await self.claude_model.get_slack_messages()
last_msg = slack_msgs[-1]["text"] if slack_msgs and len(slack_msgs) > 0 else ""
last_msg = (
slack_msgs[-1]["text"]
if slack_msgs and len(slack_msgs) > 0
else ""
)
if last_msg:
self.child.send(last_msg)
print('-------- receive final ---------')
self.child.send('[Finish]')
print("-------- receive final ---------")
self.child.send("[Finish]")
def run(self):
"""
这个函数运行在子进程
@@ -146,22 +161,24 @@ class ClaudeHandle(Process):
self.local_history = []
if (self.claude_model is None) or (not self.success):
# 代理设置
proxies = get_conf('proxies')
proxies = get_conf("proxies")
if proxies is None:
self.proxies_https = None
else:
self.proxies_https = proxies['https']
self.proxies_https = proxies["https"]
try:
SLACK_CLAUDE_USER_TOKEN = get_conf('SLACK_CLAUDE_USER_TOKEN')
self.claude_model = SlackClient(token=SLACK_CLAUDE_USER_TOKEN, proxy=self.proxies_https)
print('Claude组件初始化成功。')
SLACK_CLAUDE_USER_TOKEN = get_conf("SLACK_CLAUDE_USER_TOKEN")
self.claude_model = SlackClient(
token=SLACK_CLAUDE_USER_TOKEN, proxy=self.proxies_https
)
print("Claude组件初始化成功。")
except:
self.success = False
tb_str = '\n```\n' + trimmed_format_exc() + '\n```\n'
self.child.send(f'[Local Message] 不能加载Claude组件。{tb_str}')
self.child.send('[Fail]')
self.child.send('[Finish]')
tb_str = "\n```\n" + trimmed_format_exc() + "\n```\n"
self.child.send(f"[Local Message] 不能加载Claude组件。{tb_str}")
self.child.send("[Fail]")
self.child.send("[Finish]")
raise RuntimeError(f"不能加载Claude组件。")
self.success = True
@@ -169,42 +186,49 @@ class ClaudeHandle(Process):
# 进入任务等待状态
asyncio.run(self.async_run())
except Exception:
tb_str = '\n```\n' + trimmed_format_exc() + '\n```\n'
self.child.send(f'[Local Message] Claude失败 {tb_str}.')
self.child.send('[Fail]')
self.child.send('[Finish]')
tb_str = "\n```\n" + trimmed_format_exc() + "\n```\n"
self.child.send(f"[Local Message] Claude失败 {tb_str}.")
self.child.send("[Fail]")
self.child.send("[Finish]")
def stream_chat(self, **kwargs):
"""
这个函数运行在主进程
"""
self.threadLock.acquire()
self.parent.send(kwargs) # 发送请求到子进程
self.parent.send(kwargs) # 发送请求到子进程
while True:
res = self.parent.recv() # 等待Claude回复的片段
if res == '[Finish]':
break # 结束
elif res == '[Fail]':
res = self.parent.recv() # 等待Claude回复的片段
if res == "[Finish]":
break # 结束
elif res == "[Fail]":
self.success = False
break
else:
yield res # Claude回复的片段
yield res # Claude回复的片段
self.threadLock.release()
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第三部分：主进程统一调用函数接口
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
global claude_handle
claude_handle = None
def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="", observe_window=None, console_slience=False):
def predict_no_ui_long_connection(
inputs,
llm_kwargs,
history=[],
sys_prompt="",
observe_window=None,
console_slience=False,
):
"""
多线程方法
函数的说明请见 request_llms/bridge_all.py
多线程方法
函数的说明请见 request_llms/bridge_all.py
"""
global claude_handle
if (claude_handle is None) or (not claude_handle.success):
@@ -217,24 +241,40 @@ def predict_no_ui_long_connection(inputs, llm_kwargs, history=[], sys_prompt="",
# 没有 sys_prompt 接口，因此把prompt加入 history
history_feedin = []
for i in range(len(history)//2):
history_feedin.append([history[2*i], history[2*i+1]])
for i in range(len(history) // 2):
history_feedin.append([history[2 * i], history[2 * i + 1]])
watch_dog_patience = 5 # 看门狗 (watchdog) 的耐心, 设置5秒即可
response = ""
observe_window[0] = "[Local Message] 等待Claude响应中 ..."
for response in claude_handle.stream_chat(query=inputs, history=history_feedin, system_prompt=sys_prompt, max_length=llm_kwargs['max_length'], top_p=llm_kwargs['top_p'], temperature=llm_kwargs['temperature']):
for response in claude_handle.stream_chat(
query=inputs,
history=history_feedin,
system_prompt=sys_prompt,
max_length=llm_kwargs["max_length"],
top_p=llm_kwargs["top_p"],
temperature=llm_kwargs["temperature"],
):
observe_window[0] = preprocess_newbing_out_simple(response)
if len(observe_window) >= 2:
if (time.time()-observe_window[1]) > watch_dog_patience:
if (time.time() - observe_window[1]) > watch_dog_patience:
raise RuntimeError("程序终止。")
return preprocess_newbing_out_simple(response)
def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_prompt='', stream=True, additional_fn=None):
def predict(
inputs,
llm_kwargs,
plugin_kwargs,
chatbot,
history=[],
system_prompt="",
stream=True,
additional_fn=None,
):
"""
单线程方法
函数的说明请见 request_llms/bridge_all.py
单线程方法
函数的说明请见 request_llms/bridge_all.py
"""
chatbot.append((inputs, "[Local Message] 等待Claude响应中 ..."))
@@ -249,21 +289,30 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
if additional_fn is not None:
from core_functional import handle_core_functionality
inputs, history = handle_core_functionality(additional_fn, inputs, history, chatbot)
inputs, history = handle_core_functionality(
additional_fn, inputs, history, chatbot
)
history_feedin = []
for i in range(len(history)//2):
history_feedin.append([history[2*i], history[2*i+1]])
for i in range(len(history) // 2):
history_feedin.append([history[2 * i], history[2 * i + 1]])
chatbot[-1] = (inputs, "[Local Message] 等待Claude响应中 ...")
response = "[Local Message] 等待Claude响应中 ..."
yield from update_ui(chatbot=chatbot, history=history, msg="Claude响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。")
for response in claude_handle.stream_chat(query=inputs, history=history_feedin, system_prompt=system_prompt):
yield from update_ui(
chatbot=chatbot, history=history, msg="Claude响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。"
)
for response in claude_handle.stream_chat(
query=inputs, history=history_feedin, system_prompt=system_prompt
):
chatbot[-1] = (inputs, preprocess_newbing_out(response))
yield from update_ui(chatbot=chatbot, history=history, msg="Claude响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。")
yield from update_ui(
chatbot=chatbot, history=history, msg="Claude响应缓慢，尚未完成全部响应，请耐心完成后再提交新问题。"
)
if response == "[Local Message] 等待Claude响应中 ...":
response = "[Local Message] Claude响应异常，请刷新界面重试 ..."
history.extend([inputs, response])
logging.info(f'[raw_input] {inputs}')
logging.info(f'[response] {response}')
logging.info(f"[raw_input] {inputs}")
logging.info(f"[response] {response}")
yield from update_ui(chatbot=chatbot, history=history, msg="完成全部响应，请提交新问题。")

2
request_llms/bridge_zhipu.py
View File

@@ -42,7 +42,7 @@ def predict(inputs, llm_kwargs, plugin_kwargs, chatbot, history=[], system_promp
try:
check_packages(["zhipuai"])
except:
yield from update_ui_lastest_msg(f"导入软件依赖失败。使用该模型需要额外依赖，安装方法```pip install --upgrade zhipuai```。",
yield from update_ui_lastest_msg(f"导入软件依赖失败。使用该模型需要额外依赖，安装方法```pip install zhipuai==1.0.7```。",
chatbot=chatbot, history=history, delay=0)
return

229
request_llms/com_google.py Normal file
View File

@@ -0,0 +1,229 @@
# encoding: utf-8
# @Time : 2023/12/25
# @Author : Spike
# @Descr :
import json
import os
import re
import requests
from typing import List, Dict, Tuple
from toolbox import get_conf, encode_image, get_pictures_list
proxies, TIMEOUT_SECONDS = get_conf("proxies", "TIMEOUT_SECONDS")
"""
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第五部分 一些文件处理方法
files_filter_handler 根据type过滤文件
input_encode_handler 提取input中的文件，并解析
file_manifest_filter_html 根据type过滤文件, 并解析为html or md 文本
link_mtime_to_md 文件增加本地时间参数，避免下载到缓存文件
html_view_blank 超链接
html_local_file 本地文件取相对路径
to_markdown_tabs 文件list 转换为 md tab
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
def files_filter_handler(file_list):
new_list = []
filter_ = [
"png",
"jpg",
"jpeg",
"bmp",
"svg",
"webp",
"ico",
"tif",
"tiff",
"raw",
"eps",
]
for file in file_list:
file = str(file).replace("file=", "")
if os.path.exists(file):
if str(os.path.basename(file)).split(".")[-1] in filter_:
new_list.append(file)
return new_list
def input_encode_handler(inputs, llm_kwargs):
if llm_kwargs["most_recent_uploaded"].get("path"):
image_paths = get_pictures_list(llm_kwargs["most_recent_uploaded"]["path"])
md_encode = []
for md_path in image_paths:
type_ = os.path.splitext(md_path)[1].replace(".", "")
type_ = "jpeg" if type_ == "jpg" else type_
md_encode.append({"data": encode_image(md_path), "type": type_})
return inputs, md_encode
def file_manifest_filter_html(file_list, filter_: list = None, md_type=False):
new_list = []
if not filter_:
filter_ = [
"png",
"jpg",
"jpeg",
"bmp",
"svg",
"webp",
"ico",
"tif",
"tiff",
"raw",
"eps",
]
for file in file_list:
if str(os.path.basename(file)).split(".")[-1] in filter_:
new_list.append(html_local_img(file, md=md_type))
elif os.path.exists(file):
new_list.append(link_mtime_to_md(file))
else:
new_list.append(file)
return new_list
def link_mtime_to_md(file):
link_local = html_local_file(file)
link_name = os.path.basename(file)
a = f"[{link_name}]({link_local}?{os.path.getmtime(file)})"
return a
def html_local_file(file):
base_path = os.path.dirname(__file__) # 项目目录
if os.path.exists(str(file)):
file = f'file={file.replace(base_path, ".")}'
return file
def html_local_img(__file, layout="left", max_width=None, max_height=None, md=True):
style = ""
if max_width is not None:
style += f"max-width: {max_width};"
if max_height is not None:
style += f"max-height: {max_height};"
__file = html_local_file(__file)
a = f'<div align="{layout}"><img src="{__file}" style="{style}"></div>'
if md:
a = f"![{__file}]({__file})"
return a
def to_markdown_tabs(head: list, tabs: list, alignment=":---:", column=False):
"""
Args:
head: 表头：[]
tabs: 表值：[[列1], [列2], [列3], [列4]]
alignment: :--- 左对齐， :---: 居中对齐， ---: 右对齐
column: True to keep data in columns, False to keep data in rows (default).
Returns:
A string representation of the markdown table.
"""
if column:
transposed_tabs = list(map(list, zip(*tabs)))
else:
transposed_tabs = tabs
# Find the maximum length among the columns
max_len = max(len(column) for column in transposed_tabs)
tab_format = "| %s "
tabs_list = "".join([tab_format % i for i in head]) + "|\n"
tabs_list += "".join([tab_format % alignment for i in head]) + "|\n"
for i in range(max_len):
row_data = [tab[i] if i < len(tab) else "" for tab in transposed_tabs]
row_data = file_manifest_filter_html(row_data, filter_=None)
tabs_list += "".join([tab_format % i for i in row_data]) + "|\n"
return tabs_list
class GoogleChatInit:
def __init__(self):
self.url_gemini = "https://generativelanguage.googleapis.com/v1beta/models/%m:streamGenerateContent?key=%k"
def generate_chat(self, inputs, llm_kwargs, history, system_prompt):
headers, payload = self.generate_message_payload(
inputs, llm_kwargs, history, system_prompt
)
response = requests.post(
url=self.url_gemini,
headers=headers,
data=json.dumps(payload),
stream=True,
proxies=proxies,
timeout=TIMEOUT_SECONDS,
)
return response.iter_lines()
def __conversation_user(self, user_input, llm_kwargs):
what_i_have_asked = {"role": "user", "parts": []}
if "vision" not in self.url_gemini:
input_ = user_input
encode_img = []
else:
input_, encode_img = input_encode_handler(user_input, llm_kwargs=llm_kwargs)
what_i_have_asked["parts"].append({"text": input_})
if encode_img:
for data in encode_img:
what_i_have_asked["parts"].append(
{
"inline_data": {
"mime_type": f"image/{data['type']}",
"data": data["data"],
}
}
)
return what_i_have_asked
def __conversation_history(self, history, llm_kwargs):
messages = []
conversation_cnt = len(history) // 2
if conversation_cnt:
for index in range(0, 2 * conversation_cnt, 2):
what_i_have_asked = self.__conversation_user(history[index], llm_kwargs)
what_gpt_answer = {
"role": "model",
"parts": [{"text": history[index + 1]}],
}
messages.append(what_i_have_asked)
messages.append(what_gpt_answer)
return messages
def generate_message_payload(
self, inputs, llm_kwargs, history, system_prompt
) -> Tuple[Dict, Dict]:
messages = [
# {"role": "system", "parts": [{"text": system_prompt}]}, # gemini 不允许对话轮次为偶数，所以这个没有用，看后续支持吧。。。
# {"role": "user", "parts": [{"text": ""}]},
# {"role": "model", "parts": [{"text": ""}]}
]
self.url_gemini = self.url_gemini.replace(
"%m", llm_kwargs["llm_model"]
).replace("%k", get_conf("GEMINI_API_KEY"))
header = {"Content-Type": "application/json"}
if "vision" not in self.url_gemini: # 不是vision 才处理history
messages.extend(
self.__conversation_history(history, llm_kwargs)
) # 处理 history
messages.append(self.__conversation_user(inputs, llm_kwargs)) # 处理用户对话
payload = {
"contents": messages,
"generationConfig": {
# "maxOutputTokens": 800,
"stopSequences": str(llm_kwargs.get("stop", "")).split(" "),
"temperature": llm_kwargs.get("temperature", 1),
"topP": llm_kwargs.get("top_p", 0.8),
"topK": 10,
},
}
return header, payload
if __name__ == "__main__":
google = GoogleChatInit()
# print(gootle.generate_message_payload('你好呀', {}, ['123123', '3123123'], ''))
# gootle.input_encode_handle('123123[123123](./123123), ![53425](./asfafa/fff.jpg)')

95
request_llms/com_skylark2api.py Normal file
View File

@@ -0,0 +1,95 @@
from toolbox import get_conf
import threading
import logging
import os
timeout_bot_msg = '[Local Message] Request timeout. Network error.'
#os.environ['VOLC_ACCESSKEY'] = ''
#os.environ['VOLC_SECRETKEY'] = ''
class YUNQUERequestInstance():
def __init__(self):
self.time_to_yield_event = threading.Event()
self.time_to_exit_event = threading.Event()
self.result_buf = ""
def generate(self, inputs, llm_kwargs, history, system_prompt):
# import _thread as thread
from volcengine.maas import MaasService, MaasException
maas = MaasService('maas-api.ml-platform-cn-beijing.volces.com', 'cn-beijing')
YUNQUE_SECRET_KEY, YUNQUE_ACCESS_KEY,YUNQUE_MODEL = get_conf("YUNQUE_SECRET_KEY", "YUNQUE_ACCESS_KEY","YUNQUE_MODEL")
maas.set_ak(YUNQUE_ACCESS_KEY) #填写 VOLC_ACCESSKEY
maas.set_sk(YUNQUE_SECRET_KEY) #填写 'VOLC_SECRETKEY'
self.result_buf = ""
req = {
"model": {
"name": YUNQUE_MODEL,
"version": "1.0", # use default version if not specified.
},
"parameters": {
"max_new_tokens": 4000, # 输出文本的最大tokens限制
"min_new_tokens": 1, # 输出文本的最小tokens限制
"temperature": llm_kwargs['temperature'], # 用于控制生成文本的随机性和创造性，Temperature值越大随机性越大，取值范围0~1
"top_p": llm_kwargs['top_p'], # 用于控制输出tokens的多样性，TopP值越大输出的tokens类型越丰富，取值范围0~1
"top_k": 0, # 选择预测值最大的k个token进行采样，取值范围0-1000，0表示不生效
"max_prompt_tokens": 4000, # 最大输入 token 数，如果给出的 prompt 的 token 长度超过此限制，取最后 max_prompt_tokens 个 token 输入模型。
},
"messages": self.generate_message_payload(inputs, llm_kwargs, history, system_prompt)
}
response = maas.stream_chat(req)
for resp in response:
self.result_buf += resp.choice.message.content
yield self.result_buf
'''
for event in response.events():
if event.event == "add":
self.result_buf += event.data
yield self.result_buf
elif event.event == "error" or event.event == "interrupted":
raise RuntimeError("Unknown error:" + event.data)
elif event.event == "finish":
yield self.result_buf
break
else:
raise RuntimeError("Unknown error:" + str(event))
logging.info(f'[raw_input] {inputs}')
logging.info(f'[response] {self.result_buf}')
'''
return self.result_buf
def generate_message_payload(inputs, llm_kwargs, history, system_prompt):
from volcengine.maas import ChatRole
conversation_cnt = len(history) // 2
messages = [{"role": ChatRole.USER, "content": system_prompt},
{"role": ChatRole.ASSISTANT, "content": "Certainly!"}]
if conversation_cnt:
for index in range(0, 2 * conversation_cnt, 2):
what_i_have_asked = {}
what_i_have_asked["role"] = ChatRole.USER
what_i_have_asked["content"] = history[index]
what_gpt_answer = {}
what_gpt_answer["role"] = ChatRole.ASSISTANT
what_gpt_answer["content"] = history[index + 1]
if what_i_have_asked["content"] != "":
if what_gpt_answer["content"] == "":
continue
if what_gpt_answer["content"] == timeout_bot_msg:
continue
messages.append(what_i_have_asked)
messages.append(what_gpt_answer)
else:
messages[-1]['content'] = what_gpt_answer['content']
what_i_ask_now = {}
what_i_ask_now["role"] = ChatRole.USER
what_i_ask_now["content"] = inputs
messages.append(what_i_ask_now)
return messages

9
request_llms/com_zhipuapi.py
View File

@@ -21,11 +21,13 @@ class ZhipuRequestInstance():
response = zhipuai.model_api.sse_invoke(
model=ZHIPUAI_MODEL,
prompt=generate_message_payload(inputs, llm_kwargs, history, system_prompt),
top_p=llm_kwargs['top_p'],
temperature=llm_kwargs['temperature'],
top_p=llm_kwargs['top_p']*0.7, # 智谱的API抽风，手动*0.7给做个线性变换
temperature=llm_kwargs['temperature']*0.95, # 智谱的API抽风，手动*0.7给做个线性变换
)
for event in response.events():
if event.event == "add":
# if self.result_buf == "" and event.data.startswith(" "):
# event.data = event.data.lstrip(" ") # 每次智谱为啥都要带个空格开头呢？
self.result_buf += event.data
yield self.result_buf
elif event.event == "error" or event.event == "interrupted":
@@ -35,7 +37,8 @@ class ZhipuRequestInstance():
break
else:
raise RuntimeError("Unknown error:" + str(event))
if self.result_buf == "":
yield "智谱没有返回任何数据, 请检查ZHIPUAI_API_KEY和ZHIPUAI_MODEL是否填写正确."
logging.info(f'[raw_input] {inputs}')
logging.info(f'[response] {self.result_buf}')
return self.result_buf

52
request_llms/edge_gpt_free.py
View File

@@ -1,8 +1,8 @@
"""
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
第一部分：来自EdgeGPT.py
https://github.com/acheong08/EdgeGPT
========================================================================
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
"""
"""
Main.py
@@ -196,9 +196,9 @@ class _ChatHubRequest:
self,
prompt: str,
conversation_style: CONVERSATION_STYLE_TYPE,
options = None,
webpage_context = None,
search_result = False,
options=None,
webpage_context=None,
search_result=False,
) -> None:
"""
Updates request object
@@ -294,9 +294,9 @@ class _Conversation:
def __init__(
self,
proxy = None,
async_mode = False,
cookies = None,
proxy=None,
async_mode=False,
cookies=None,
) -> None:
if async_mode:
return
@@ -350,8 +350,8 @@ class _Conversation:
@staticmethod
async def create(
proxy = None,
cookies = None,
proxy=None,
cookies=None,
):
self = _Conversation(async_mode=True)
self.struct = {
@@ -418,8 +418,8 @@ class _ChatHub:
def __init__(
self,
conversation: _Conversation,
proxy = None,
cookies = None,
proxy=None,
cookies=None,
) -> None:
self.session = None
self.wss = None
@@ -441,7 +441,7 @@ class _ChatHub:
conversation_style: CONVERSATION_STYLE_TYPE = None,
raw: bool = False,
options: dict = None,
webpage_context = None,
webpage_context=None,
search_result: bool = False,
) -> Generator[str, None, None]:
"""
@@ -452,10 +452,12 @@ class _ChatHub:
ws_cookies = []
for cookie in self.cookies:
ws_cookies.append(f"{cookie['name']}={cookie['value']}")
req_header.update({
'Cookie': ';'.join(ws_cookies),
})
req_header.update(
{
"Cookie": ";".join(ws_cookies),
}
)
timeout = aiohttp.ClientTimeout(total=30)
self.session = aiohttp.ClientSession(timeout=timeout)
@@ -521,9 +523,9 @@ class _ChatHub:
msg = await self.wss.receive()
try:
objects = msg.data.split(DELIMITER)
except :
except:
continue
for obj in objects:
if obj is None or not obj:
continue
@@ -624,8 +626,8 @@ class Chatbot:
def __init__(
self,
proxy = None,
cookies = None,
proxy=None,
cookies=None,
) -> None:
self.proxy = proxy
self.chat_hub: _ChatHub = _ChatHub(
@@ -636,8 +638,8 @@ class Chatbot:
@staticmethod
async def create(
proxy = None,
cookies = None,
proxy=None,
cookies=None,
):
self = Chatbot.__new__(Chatbot)
self.proxy = proxy
@@ -654,7 +656,7 @@ class Chatbot:
wss_link: str = "wss://sydney.bing.com/sydney/ChatHub",
conversation_style: CONVERSATION_STYLE_TYPE = None,
options: dict = None,
webpage_context = None,
webpage_context=None,
search_result: bool = False,
) -> dict:
"""
@@ -680,7 +682,7 @@ class Chatbot:
conversation_style: CONVERSATION_STYLE_TYPE = None,
raw: bool = False,
options: dict = None,
webpage_context = None,
webpage_context=None,
search_result: bool = False,
) -> Generator[str, None, None]:
"""

2
request_llms/requirements_chatglm.txt
View File

@@ -2,4 +2,4 @@ protobuf
cpm_kernels
torch>=1.10
mdtex2html
sentencepiece
sentencepiece

2
request_llms/requirements_jittorllms.txt
View File

@@ -3,4 +3,4 @@ jtorch >= 0.1.3
torch
torchvision
pandas
jieba
jieba

1
request_llms/requirements_moss.txt
View File

@@ -5,4 +5,3 @@ accelerate
matplotlib
huggingface_hub
triton

2
request_llms/requirements_qwen.txt
View File

@@ -1 +1 @@
dashscope
dashscope

2
request_llms/requirements_qwen_local.txt
View File

@@ -2,4 +2,4 @@ modelscope
transformers_stream_generator
auto-gptq
optimum
urllib3<2
urllib3<2

2
request_llms/requirements_slackclaude.txt
View File

@@ -1 +1 @@
slack-sdk==3.21.3
slack-sdk==3.21.3