Merge branch 'v3.1' of github.com:binary-husky/chatgpt_academic into v3.1

README.md

@@ -1,4 +1,6 @@
-# ChatGPT 学术优化
+
+
+# <img src="docs/logo.png" width="40" > ChatGPT 学术优化
 
 **如果喜欢这个项目，请给它一个Star；如果你发明了更好用的快捷键或函数插件，欢迎发issue或者pull requests**
 
@@ -68,11 +70,13 @@ huggingface免科学上网[在线体验](https://huggingface.co/spaces/qingxu98/
 <img src="https://user-images.githubusercontent.com/96192199/226935232-6b6a73ce-8900-4aee-93f9-733c7e6fef53.png" width="700" >
 </div>
 
-- 多种大语言模型混合调用（[v3.1分支](https://github.com/binary-husky/chatgpt_academic/tree/v3.1)测试中）
+- 多种大语言模型混合调用（ChatGLM + OpenAI-GPT3.5 + [API2D](https://api2d.com/)-GPT4， [v3.1分支](https://github.com/binary-husky/chatgpt_academic/tree/v3.1)测试中）
 <div align="center">
-<img src="https://user-images.githubusercontent.com/96192199/231222778-34776885-a7f0-4f2c-b5f4-7cc2ef3ecb58.png" width="700" >
+<img src="https://user-images.githubusercontent.com/96192199/232537274-deca0563-7aa6-4b5d-94a2-b7c453c47794.png" width="700" >
 </div>
 
+v3.1的[huggingface测试版](https://huggingface.co/spaces/qingxu98/academic-chatgpt-beta)（huggingface版不支持chatglm）
+
 
 
 ## 直接运行 (Windows, Linux or MacOS)

crazy_functions/crazy_utils.py

@@ -448,6 +448,7 @@ def read_and_clean_pdf_text(fp):
                     pf = 998
                     for l in t['lines']:
                         txt_line = "".join([wtf['text'] for wtf in l['spans']])
+                        if len(txt_line) == 0: continue
                         pf = primary_ffsize(l)
                         meta_line.append([txt_line, pf, l['bbox'], l])
                         for wtf in l['spans']: # for l in t['lines']:
@@ -558,8 +559,8 @@ def read_and_clean_pdf_text(fp):
         meta_txt = meta_txt.replace('\n', '\n\n')
 
         ############################## <第 5 步，展示分割效果> ##################################
-        for f in finals:
-            print亮黄(f)
-            print亮绿('***************************')
+        # for f in finals:
+        #    print亮黄(f)
+        #    print亮绿('***************************')
 
     return meta_txt, page_one_meta

crazy_functions/批量翻译PDF文档_多线程.py

@@ -13,7 +13,7 @@ def 批量翻译PDF文档(txt, llm_kwargs, plugin_kwargs, chatbot, history, sys_
     # 基本信息：功能、贡献者
     chatbot.append([
         "函数插件功能？",
-        "批量总结PDF文档。函数插件贡献者: Binary-Husky"])
+        "批量翻译PDF文档。函数插件贡献者: Binary-Husky"])
     yield from update_ui(chatbot=chatbot, history=history) # 刷新界面
 
     # 尝试导入依赖，如果缺少依赖，则给出安装建议
@@ -59,7 +59,7 @@ def 批量翻译PDF文档(txt, llm_kwargs, plugin_kwargs, chatbot, history, sys_
 def 解析PDF(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatbot, history, sys_prompt):
     import os
     import tiktoken
-    TOKEN_LIMIT_PER_FRAGMENT = 1600
+    TOKEN_LIMIT_PER_FRAGMENT = 1280
     generated_conclusion_files = []
     for index, fp in enumerate(file_manifest):
 
@@ -91,13 +91,13 @@ def 解析PDF(file_manifest, project_folder, llm_kwargs, plugin_kwargs, chatbot,
         # 多线，翻译
         gpt_response_collection = yield from request_gpt_model_multi_threads_with_very_awesome_ui_and_high_efficiency(
             inputs_array=[
-                f"以下是你需要翻译的论文片段：\n{frag}" for frag in paper_fragments],
+                f"你需要翻译以下内容：\n{frag}" for frag in paper_fragments],
             inputs_show_user_array=[f"\n---\n 原文： \n\n {frag.replace('#', '')}  \n---\n 翻译：\n " for frag in paper_fragments],
             llm_kwargs=llm_kwargs,
             chatbot=chatbot,
             history_array=[[paper_meta] for _ in paper_fragments],
             sys_prompt_array=[
-                "请你作为一个学术翻译，负责把学术论文的片段准确翻译成中文。" for _ in paper_fragments],
+                "请你作为一个学术翻译，负责把学术论文准确翻译成中文。注意文章中的每一句话都要翻译。" for _ in paper_fragments],
             # max_workers=5  # OpenAI所允许的最大并行过载
         )
 

docs/Dockerfile+ChatGLM

@@ -36,6 +36,9 @@ from transformers import AutoModel, AutoTokenizer \n\
 chatglm_tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True) \n\
 chatglm_model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).float() ' >> warm_up_chatglm.py
 RUN python3 -u warm_up_chatglm.py
+
+# 禁用缓存，确保更新代码
+ADD "https://www.random.org/cgi-bin/randbyte?nbytes=10&format=h" skipcache
 RUN $useProxyNetwork git pull
 
 # 为chatgpt-academic配置代理和API-KEY （非必要 可选步骤）

main.py

@@ -182,7 +182,7 @@ def main():
         threading.Thread(target=auto_update, name="self-upgrade", daemon=True).start()
 
     auto_opentab_delay()
-    demo.queue(concurrency_count=CONCURRENT_COUNT).launch(server_name="0.0.0.0", server_port=PORT, auth=AUTHENTICATION)
+    demo.queue(concurrency_count=CONCURRENT_COUNT).launch(server_name="0.0.0.0", server_port=PORT, auth=AUTHENTICATION, favicon_path="docs/logo.png")
 
 if __name__ == "__main__":
     main()

request_llm/bridge_all.py

@@ -23,6 +23,9 @@ from .bridge_tgui import predict as tgui_ui
 
 colors = ['#FF00FF', '#00FFFF', '#FF0000', '#990099', '#009999', '#990044']
 
+get_token_num_gpt35 = lambda txt: len(tiktoken.encoding_for_model("gpt-3.5-turbo").encode(txt, disallowed_special=()))
+get_token_num_gpt4 = lambda txt: len(tiktoken.encoding_for_model("gpt-4").encode(txt, disallowed_special=()))
+
 model_info = {
     # openai
     "gpt-3.5-turbo": {
@@ -31,7 +34,7 @@ model_info = {
         "endpoint": "https://api.openai.com/v1/chat/completions",
         "max_token": 4096,
         "tokenizer": tiktoken.encoding_for_model("gpt-3.5-turbo"),
-        "token_cnt": lambda txt: len(tiktoken.encoding_for_model("gpt-3.5-turbo").encode(txt, disallowed_special=())),
+        "token_cnt": get_token_num_gpt35,
     },
 
     "gpt-4": {
@@ -40,7 +43,7 @@ model_info = {
         "endpoint": "https://api.openai.com/v1/chat/completions",
         "max_token": 8192,
         "tokenizer": tiktoken.encoding_for_model("gpt-4"),
-        "token_cnt": lambda txt: len(tiktoken.encoding_for_model("gpt-4").encode(txt, disallowed_special=())),
+        "token_cnt": get_token_num_gpt4,
     },
 
     # api_2d
@@ -50,7 +53,7 @@ model_info = {
         "endpoint": "https://openai.api2d.net/v1/chat/completions",
         "max_token": 4096,
         "tokenizer": tiktoken.encoding_for_model("gpt-3.5-turbo"),
-        "token_cnt": lambda txt: len(tiktoken.encoding_for_model("gpt-3.5-turbo").encode(txt, disallowed_special=())),
+        "token_cnt": get_token_num_gpt35,
     },
 
     "api2d-gpt-4": {
@@ -59,7 +62,7 @@ model_info = {
         "endpoint": "https://openai.api2d.net/v1/chat/completions",
         "max_token": 8192,
         "tokenizer": tiktoken.encoding_for_model("gpt-4"),
-        "token_cnt": lambda txt: len(tiktoken.encoding_for_model("gpt-4").encode(txt, disallowed_special=())),
+        "token_cnt": get_token_num_gpt4,
     },
 
     # chatglm
@@ -69,7 +72,7 @@ model_info = {
         "endpoint": None,
         "max_token": 1024,
         "tokenizer": tiktoken.encoding_for_model("gpt-3.5-turbo"),
-        "token_cnt": lambda txt: len(tiktoken.encoding_for_model("gpt-3.5-turbo").encode(txt, disallowed_special=())),
+        "token_cnt": get_token_num_gpt35,
     },
 
 }

toolbox.py

@@ -401,7 +401,7 @@ def on_file_uploaded(files, chatbot, txt, txt2, checkboxes):
     chatbot.append(['我上传了文件，请查收',
                     f'[Local Message] 收到以下文件: \n\n{moved_files_str}' +
                     f'\n\n调用路径参数已自动修正到: \n\n{txt}' +
-                    f'\n\n现在您点击任意实验功能时，以上文件将被作为输入参数'+err_msg])
+                    f'\n\n现在您点击任意“红颜色”标识的函数插件时，以上文件将被作为输入参数'+err_msg])
     return chatbot, txt, txt2